You are on page 1of 254

1

MAKE MY DAY: RITUAL, DEPENDENCY AND THE HABIT OF NEWSPAPER READING

by CLYDE H. BENTLEY

A DISSERTATION Presented to the School of Journalism and Communication and the Graduate School of the University of Oregon partial fulfillment of the requirements for the degree of Doctor of Philosophy June 2000

"Make My Day: Ritual, Dependency and the Habit of Newspaper Reading" is a dissertation prepared by Clyde H. Bentley in partial fulfillment of the requirements for the Doctor of Philosophy degree in the School of Journalism and Communication. This dissertation has been approved and accepted by:

Dr. John Russial, Chair of the Examining Committee

Date

Committee in charge:

Dr. Dr. Dr. Dr.

John Russial, Chair Debra Merskin Arnold Ismach Lynn Kahle

Accepted by:

Dean of the Graduate School

2000 Clyde H. Bentley

An Abstract of the Dissertation of Clyde Howard Bentley for the degree of Doctor of Philosophy

in the School of Journalism and Communication to be taken June 2000

Title: MAKE MY DAY: RITUAL, DEPENDENCY AND THE HABIT OF NEWSPAPER READING

Approved:__________________________________ Dr. John Russial

Newspapers entered the 21st century as a continuing enigma. Competition for the attention of readers is at an all-time high as the Internet joins television, radio, magazines and other media in providing news, entertainment and information. But while alternative media can

and do provide information faster, more frequently and cheaper than newspapers, this medium that is little changed from its 18th century debut remains enormously popular. This dissertation explores the power of habit and ritual on newspaper readership. While some newspaper habits such as the morning trudge down the driveway for the paper or the mind-clearing daily battle with the crossword puzzle have taken on folkloric status in popular literature, scholarly research on habitual reading is scarce. This

5 study attempts to quantify the so-called "newspaper habit" and to measure its impact on readership. Data from two randomized telephone surveys of Oregon residents, one conducted in 1998 and the other in 1999, are used to accomplish this goal. Both asked questions about the intensity and logistics of newspaper use that allowed the construction of a newspaper reading habit index. The first survey also asked how readers related newspaper reading to their sense of community and related social constructs. The second survey also catalogued readership of various sections of the newspaper. In addition, it expanded on the original survey by using new criteria for newspaper readership and by probing for emotional indicators of reading habit. Data analysis revealed that habit is a force in newspaper readership that is independent of the traditional demographic drivers of readership. Regression analysis clearly showed that while the effect of habit on both the perceived importance of reading a newspaper and on reading frequency is small, it is significant, very definite and quite robust. Readers with high levels of newspaper habit also have identifiable characteristics. They were less concerned with utility than readers with low levels of habit and were more likely to feel "inconvenienced" if newspaper delivery is delayed. Entertainment-related items such as comics and crosswords were of greater interest to those with high levels of habit, but classified advertising was of lower interest.

CURRICULUM VITA

NAME OF AUTHOR: Clyde Howard Bentley PLACE OF BIRTH: Frankfurt, Germany DATE OF BIRTH: March 9, 1951

GRADUATE AND UNDERGRADUATE SCHOOLS ATTENDED: University of Oregon University of Texas Pepperdine University Shasta College

DEGREES AWARDED: Doctor of Philosophy in Journalism, 2000, University of Oregon Master of Arts in Journalism, 1990, University of Texas Bachelor of Arts in Journalism, 1973, Pepperdine University Associate of Arts in Journalism, 1971, Shasta College

AREAS OF SPECIAL INTEREST: Audience Studies Media Management

PROFESSIONAL EXPERIENCE: General Manager, The East Oregonian, Pendleton, Oregon, 19931997. Advertising Manager, The Irving News, Irving, Texas, 1991-1993 Marketing Director, The Recorder-Times, San Antonio, Texas,

7 1990-1991 Managing Editor, The Coeur d'Alene Press, Coeur d'Alene, Idaho, 1981-1988 Assistant to the Publisher, The Headlight-Herald, Tillamook, Oregon, 1980-1981 Editor, The Tahoe World, Tahoe City, California, 1978-1979 Managing Editor, The Redding Outlook, Redding, California, 19761978 Reporter, The Record-Searchlight, Redding, California 1973-1976 Intern, Newsweek Magazine, Los Angeles Bureau, Los Angeles, California, 1972-1973 Reporter, The Community Informer, Los Angeles, California, 19721972

AWARDS AND HONORS: Kappa Tau Alpha, 1998-2000

PUBLICATIONS: Bentley, Clyde. 1998. What It Means to Miss the Newspaper. Ideas, September, 6(8). Stein, Andi, Clyde Bentley, and Wayne Wanta. 1998. Delivering the News: A Study of Employee Communications Practices Used by Newspaper Editors. Proceedings of the Public Relations/Corporate Communications Track of the 11th Annual Meeting of the International Academy of Business Disciplines, March 25-28, at Chicago.

ACKNOWLEGEMENTS

A single academic, working alone in his cubicle, has little hope of completing a research project such as this. Although the dissertation bears my name, it was accomplished only with the generous help and advise of countless people. My sincere thanks goes to all those who advised, prodded, edited and supported me. In particular, special thanks is owed to: John Russial, who is not only chair of my committee, but both the finest editor in academia and living proof that the best news professionals are also among the best scholars. Debra Merskin, who introduced me to Media Systems Dependency and insisted this old newsman stick to legitimate theory. Arnold Ismach, who long before he became my methodology mentor gave a speech that eventually turned this editor into a scholar. Lynn Kahle, who is a man for all seasons and all quadrants of the university. On paper, Lynn taught me about consumer behavior. In reality, he is a living lesson in academic convergence. Patricia Gwartney, director of the Oregon Survey Research Laboratory and an incredibly patient evangelist of survey methodology. Andi Stein, a fellow traveler on the doctoral road who provided moral support from the office next door.

DEDICATION

This dissertation and all that it represents is dedicated to that very special love that moves mountains, changes history, inspires poets and creates scholars. The love of my wonderful wife, Cecile, my son, Garrett and my daughter, Gillian, is each day expressed in much more than physical and emotional support. My family believed in me when I had my doubts believed that a blue-collar kid could scratch his way up through the newsroom, become an editor and eventually don the hood of an academic doctor. Their incredible accomplishments can be measured by no survey and analyzed by no statistic, but have far more power and impact than any variable this humble researcher could hope to define. To paraphrase the songwriter, this is dedicated to the ones I love.

10 TABLE OF CONTENTS Chapter I. Page INTRODUCTION.................................................................. 1 Implications of the Study .................................................7 Organization ..................................................................8 II. HISTORICAL BACKGROUND OF READER RESEARCH............ 1 0 Administrative Research ............................................... 1 4 Reading Frequency and Penetration ............................... 1 8 Theoretical Research .....................................................2 6 Convergence.................................................................2 8 III. READERSHIP THEORY.......................................................3 0 Uses and Gratification ...................................................3 4 Media System Dependency.............................................3 9 Comparison and Convergence ........................................ 4 5 Uses and Dependency ....................................................4 8 IV. HABIT AND RITUAL .......................................................... 5 2 Early Views of Habit...................................................... 5 2 Popular Literature ........................................................6 2

11 Page V. PRELIMINARY STUDIES .....................................................6 6 1990 Study: Introduction .............................................6 7 1990 Study: Hypotheses and Methodology......................7 2 1990 Study: Key Definitions..........................................7 6 1990 Study: Results .....................................................7 7 1990 Study: Review of Hypotheses.................................8 7 1990 Study: Discussion.................................................8 9 1998 Study: Introduction ............................................. 9 1 1998 Study: Berelson's Research....................................9 2 1998 Study: Research Questions and Methodology ..........9 7 1998 Study: Findings ................................................. 1 0 0 1998 Study: Discussion............................................... 1 0 6 Preliminary Studies: Discussion ................................... 1 0 8 VI. METHODOLOGY............................................................... 1 1 0 Research Questions...................................................... 1 1 1 Hypotheses................................................................. 1 1 3 Survey Overview........................................................ 1 1 5 Study I Survey Questions............................................. 1 1 8 Study II Survey Questions............................................ 1 2 0 The Study Area........................................................... 1 2 6 Limitations................................................................. 1 2 8 VII. RESULTS OF STUDY I .................................................. 1 3 0 Variables.................................................................... 1 3 0 Method of Analysis ...................................................... 1 3 4 Findings..................................................................... 1 3 5 VIII. RESULTS OF STUDY II................................................. 1 4 9 Variables.................................................................... 1 4 5 Method of Analysis ...................................................... 1 5 3 Findings..................................................................... 1 5 4 IX. DISCUSSION ................................................................... 1 7 2 Review of the Hypotheses............................................. 1 7 3 Implications ............................................................... 1 8 3

12 Page Further Research ........................................................ 1 8 9 Recommendation to Newspaper Publishers .................... 1 9 1 Conclusion.................................................................. 1 9 2 APPENDIX A. B. STUDY 1 QUESTIONS....................................................... 1 9 4 STUDY 2 QUESTIONS ...................................................... 2 1 5

BIBLIOGRAPHY...............................................................................2 2 6

13

LIST OF TABLES Table 1 Page Ball-Rokeach's Comparison of Uses and Gratifications with Media Systems ............................................................4 6 Newspaper Readership by Frequency ........................................7 8 Frequency and Percentages for Demographics Questions by Non-readers, Infrequent Readers and Frequent Readers...............................................................................8 4 Valid Sample Size Needed for Various Populations..................... 1 1 8 Oregon Comparative Statistics ............................................... 1 2 7 Newspaper Reading Habit Index Variables............................... 1 3 3 Importance of Newspaper Reading by Habit Level of Newspaper Reading Behavior .......................................... 1 3 6 Means, Standard Deviations and Correlations for Variables...................................................................... 1 3 6 B Coefficients for Logistic Regression of Importance of Regularly Reading a Newspaper....................................... 1 4 1 Unstandardized Ordinary Least Square Coefficients for Regression of Intensity of Newspaper Subscription............ 1 4 5 Frequencies for "In the Average Week, How Often Do You Normally Read a Newspaper?".................................. 1 5 0 Study 2 Newspaper Reading Habit Index Variables .......................................................................... 1 5 2 Frequencies for Media Habit Index .......................................... 1 5 4 Means, Standard Deviations and Correlations for Study 2 Variable........................................................... 1 5 5

2 3

4 5 6 7

10

11

12

13 14

14 Page 15 Unstandardized Ordinary Least Square Coefficients for Regression of Importance of Frequency of Reading a Newspaper...................................................... 1 5 7 First, Second and Third Choice of Most Important Section or Page to Read ............................................................... 1 6 1 Rankings of Most Important Section, by Level of Newspaper Reading Habit Index....................................... 1 6 4 Rankings of Most Important Section, by Level of Self-reported Newspaper Habit......................................... 1 6 5 Review of Hypotheses............................................................. 1 7 5

16

17

18

19

15

LIST OF FIGURES Figure 1 2 3 Page Newspaper Readership as a Percent of U.S. Adults.........................3 Uses and Dependency Model .....................................................4 9 Comparison of Newspaper Readers and Television Viewers...............................................................................8 0 Age and Ritual Newspaper Use................................................ 1 3 8 Education and Ritual Newspaper Use ...................................... 1 3 9 Importance of Newspaper Reading, by Age............................... 1 4 3 Habit and the Most Important Factor in Reading ...................... 1 4 7 How Do You Feel When You Are Prevented from Reading the Newspaper?..................................................... 1 6 9 Disposal: What Do You Do with an Unread Paper? .................... 1 7 0

4 5 6 7 8

16

CHAPTER I

INTRODUCTION

At some point in their careers, all newspaper editors ask the same question: Is it really just the news? Is it really only the content I supply that attracts readers to the newspaper?" Often-puzzling reader attitudes; intense loyalty to logos, comic strips and delivery times; and the vitriol some readers express about the perceived politics of a paper to which they remain loyal subscribers make many editors wonder what readers really want. What at first glance seems to be a simple question is in actuality an extremely complicated query with no simple answers but with large social and commercial implications. Although other technologies new and popular as the 19th century turned ships, printing presses, power looms, steam engines have survived, few products popularized then have remained as close to their format and popularity as have newspapers. For all its status as a late 18th century artifact, the newspaper remains impressively resilient. By definition, the very existence of any mass

17 medium depends on some sort of mass audience to consume it. In the case of the newspaper, the source of life is the reader, for whom the news is collected, the stories are crafted, the pages are printed and the finished product is tossed upon the doorstep. Newspaper readership, while rising in raw numbers over the past 50 years, has declined as a percentage of the total U.S. population (NAA 1999). Little wonder. Since its earliest days, the publishing industry has faced wave after wave of technological challenge handbills, town criers, telegraph, radio, television, audiotext, direct satellite, Internet etc. each setting off forecasts of the end of newspapers as we know them. (Rieder 1999; Walker 2000) And indeed, the challenges have taken a toll

(Figure 1). In 1950, 53.8 million newspapers circulated in the U.S. daily in a population of 152 million. In 1997, the daily circulation of newspapers had climbed to 56.7 million, but the population had soared to 268 million (NAA 1999, p. 11; U.S. Census Bureau 1998, p. 8). Despite the challenge and the hand-wringing of many publishers, the newspaper industry has remained surprisingly strong. The Newspaper Association of America estimates that 59% of American adults still pick up a weekday paper, and 69% read the Sunday paper, even though most have ample access to other forms of information (NAA 1999). Newspapers were only recently edged out by local television news as the most popularly used news medium (NAA 1998).

18

74% 72% 70% 68% 66% 64% 62%


1964 1973 1979 1983 1987 1991

TREND
1995 1999

Source: Newspaper Association of America

FIGURE 1. Newspaper Readership as a Percent of U.S. Adults

In comparison, other industries have not fared so well in the technology battle. For instance, U.S. passenger railroads during the same time period went from being the overwhelmingly dominant form of interurban transportation to a minor blip in modern airplane-andhighway-dominated transportation statistics (Martin 1995). From a pure business perspective, a product that manages to maintain near dominance despite the introduction of low cost, highly effective competitors must have either a terrific marketing program or some appealing factors not obvious at first glance. Unfortunately,

19 perhaps, the literature is clear that the newspaper industry is anything but a self-marketing wizard (Polansky and Hughes 1986; Demers 1996; Garneau 1994; Ogan, Brown, and Weaver 1979; Consoli 1995; Bentley 1998; Rockmore 1995) . Yet, the newspaper industry remains strong despite its weaknesses (NAA 1999). Within the industry, the dominant explanation for this tenacity is the perceived love of news content by newspaper readers. Although some journalists point to graphical style as the new hallmark of a good newspaper and the earliest publishers were both printers and reporters, the training and culture of journalists for 200 years has focused on content (Rieder 1999). Design and packaging have unquestionably become popular benchmarks in the past two decades, but it is the content and primarily the news content of a newspaper upon which it generally is judged in professional competition (CJR 1999). It also is the content to which newspaper managers turn for solutions when they experience problems in the marketplace. But while the adulation of news content is very strong in the newspaper journalism culture (Gladney 1995), history points to disturbing facts that call into question whether news content is the primary driver of newspaper reading. News consumers now have simply too many other options available television, radio, Internet, magazines,

20 etc. to simply "rely" upon newspapers as their only source of public information (Stempel III and Hargrove 1996). American consumers historically react to product choice by diversifying their buying patterns unless other factors intervene. (Marder 1997) This dissertation raises the possibility that the term reader may be somewhat problematic when one examines how people use newspapers. The word implies that newspaper consumers use their papers just for the information contained in them. Could there be other reasons Americans buy newspapers? Statistics imply the possibility. There is evidence that something else" besides particular content alone is at work in the attraction of Americans to newspapers, and that the "something" may be habit. This dissertation is an attempt to reveal what other factors help explain the consumer demand that has kept print-on-paper newspapers a major medium in a world of instantaneous cyber-experiences. The scenario that provided impetus to this study is from the video-screen world itself: The sit-com stereotype of Dad buried in the morning newspaper while the chaos of family life washes around him. Popular literature is replete with similar images of routines or habits linked to the newspaper: the bathrobed and beslippered neighbor slogging onto the lawn each morning for the paper, the crossword fan who cant leave for work until she finds a five-letter word for bewildered, the

21 office curmudgeon who grabs his newspaper and retires to the restroom for a half hour. This research was designed to determine whether the term the newspaper habit is more than quaint journalistic jargon. The term "habit" has in some studies been replaced with "ritual" to connote a regularly occurring activity that delivers satisfaction simply from the fact that it happens. Ritual also implies a sense of willful volition rather than biochemical addiction (as in a drug habit) (Rothenbuhler 1998). While any type of habit or desire undoubtedly has many factors, this dissertation will focus on one factor that, at first blush, seems obvious but that has received very little attention in the communications literature: the habit or ritual of newspaper reading. This research asks whether the powerful force of human habit is a significant factor in newspaper circulation. Further, this study will explore the variations of the newspaper reading habit and identify some of the contributors to that habit.

22 Implication of the Study

A study of the newspaper reading habit may have impact on both the academic field of audience studies and media industry concerns about circulation slippage. Audience studies are a relatively new specialty within communications studies (see Chapter II), but they are the focus of intense debate and research. The field has developed two major quantitative theories and two qualitative methodologies. In the quantitative school, uses and gratifications and media systems dependency have gained popularity. Qualitatively, the major audience studies areas are reception analysis and media ethnography. Nevertheless, scholars have yet to describe a unified theory for why people read the newspaper, watch television or listen to the radio. While that lofty goal is beyond the scope of this dissertation, the study will attempt to explain some of the "gaps" in audience studies and to facilitate a convergence of audience theories. From a newspaper business perspective, any insight that might prevent further deterioration of the circulation base of newspapers has significant value. U.S. newspapers are not only faced with the abovementioned decline in circulation as a proportion of national population, but an increasing problem with "churn." Churn is the phenomenon of

23 subscribing to a newspaper for a short time, then stopping the subscription only to start again at a later date. Churn at some major U.S. papers approaches 70%, which means a newspaper must replace more than twothirds of its subscribers each year (Giobbe 1994). If habit or ritual in newspaper readership can be identified and quantified, it may also be possible to encourage it. This would reduce the cost of circulation marketing to newspapers so resources could be redirected elsewhere or the net cost to the consumer and advertiser reduced.

Organization

In an effort to address the impact of habit on newspaper readership, this dissertation reviews the literature in three phases. First, a review of the history of audience studies that led to the current project is offered. Second, the dissertation reviews the theories employed in audience studies. Third, theories and research on habit and ritual are reviewed. The dissertation next reviews two preliminary studies conducted by the author that helped frame his research questions and focus his later studies. The first of these is a study of the lifestyle indicators that separate non-readers, infrequent readers and regular readers of newspapers. It was based on a telephone survey conducted in 1990. The second study was a qualitative look at newspaper subscribers whose regular delivery failed.

24 In-depth interviews of these subscribers were conducted in 1998 to determine what "missing the newspaper" means to them. The primary data upon which this dissertation is based were gathered in two omnibus telephone surveys of Oregon residents. The first survey was conducted in late 1998 and the second survey was conducted in the fall of 1999. Included in both surveys were questions about media use and habitual behavior. The final sections of the dissertation will discuss the findings of these two studies and how they relate to previous audience studies. In addition, the author will offer suggestions for the future use of the findings by other researchers and industry experts.

25

CHAPTER II

HISTORICAL BACKGROUND OF READER RESEARCH

Interest in why people read predates modern communications research and sheds some light on the roots of reader habit (Pauly 1991). The peculiar popularity and power of the early American press attracted the attention of 19th century intellectuals. Perhaps the most famous of these press observers was the French historian Alexis de Tocqueville, who was intrigued by the unofficial but critical role newspapers played in the American way of life. Unlike the autocratic European tradition, democracy American-style required common citizens to make myriad social and political decisions every day. For guidance, he noted, Americans turned to the press and incorporated newspaper reading into the routine of daily life. A newspaper is an adviser that need not be sought out, but comes of its own accord and talks to you briefly every day about the commonweal without distracting you from your private affairs (Tocqueville 1969). The 19th century newspaper reporter, philosopher and poet Walt Whitman eloquently expounded upon this theme throughout his career,

26 reminding his readers that fact was better than fiction and that factbearing newspapers were a crucial part of American life. The young country's dependence on the press was a major characteristic that distinguished it from Europe. Americans, he said, were ruled by newspapers rather than monarchs (Kaplan 1980, Page 128). This special relationship between the press and the American public would later become an element in the nation's "manifest destiny" and westward expansion. New York Tribune editor Horace Greeley tapped the

pioneer spirit he saw in all America by recycling another editor's admonition: "Go west, young man and grow up with the country." Repeated time and again, it became the mantra to action for a generation (Altschull 1990, p. 194). Greeley's vision of Americans readers and otherwise was later formalized by University of Wisconsin Professor Frederick Jackson Turner. Turner's "frontier thesis" portrayed the pioneer as the archetypal American who made democracy possible (Altschull 1990, Page 193). In 1922, Walter Lippmann revisited the American press in the essay style Tocqueville had used nearly 100 years earlier. Unlike the French

admirer of the new America, Lippmann was an insider both as an American and a journalist. The difficulty for readers, Lippmann said, is they have neither the time nor the interest to fully investigate the world outside, so they increasingly rely on the press for the pictures in their

27 heads. He referenced early quantitative research on readers to demonstrate the extent to which this social need had produced the reader habit (Lippmann 1922). Lippmann was also one of the first academics to bluntly sanctify the commodification of newspaper readers. He explained, in Public Opinion, that a penny-pinching cultural quirk among the American readership demands the use of paid advertising to facilitate the dissemination of vital information. The real problem is that the readers of a newspaper, unaccustomed to paying the cost of newsgathering, can be capitalized only by turning them into circulation that can be sold to manufacturers and merchants. And those whom it is most important to capitalize are those who have the most money to spend. Such a press is bound to respect the point of view of the buying public. It is for this buying public that newspapers are edited and published, for without that support the newspapers cannot live. (Lippmann 1922, p. 205) Newspaper audience study began to emerge into a distinctive academic pursuit around the time of World War II. That war and its predecessor had spawned great interest on the effect of mass media on the public. At first, audience studies were simply part of media effects inquiry. Eventually, however, a few scholars began examining the

media-audience relationship from the perspective of the audience itself. In 1945, Columbia University sociologist Bernard Berelson used indepth interviews in a landmark study from this perspective. Berelson and his research team interviewed New Yorkers who were deprived of their

28 daily news by an extended newspaper strike. His study was one of the first to hint at the emotional aspects of regular newspaper reading. Two other studies of the strike took approaches more in line with media effects research. The two commercial surveys focused on what alternative media people used during the strike and what sections of the newspapers they most missed. Berelson, however, intensively studied a much smaller sample to discover what people "felt" about being without a paper and what emotional or social role newspapers played in their lives. Among Berelsons findings was that fully half of those interviewed felt personally handicapped by the strike, as they had come to depend on the newspaper as a direct aid to daily life. Others pined for the respite and relaxation the newspaper provided. One of Berelsons interview subjects expressed it well: I didnt know what to do with myself. I was depressed. There was nothing to read and pass the time. I got a paper on Wednesday and felt a whole lot better. (Berelson 1949) The juxtaposition of Berelson's fascination with emotion and the other strike researchers' concentration on numbers typifies a change in the audience research field. Newspaper audience research in the post-war period quickly divided into two streams, one highly administrative and one highly theoretical.

29 Administrative Research

The macroscopic view of newspapers in daily life taken by Lippmann and Berelson took a back seat to demographic inquiries in the 1950s and 1960s. Tocqueville had been among the first scholars to imply that

newspaper readers could be categorized into groups. The success of a newspaper, he wrote, depends on attracting a like-minded group of readers into an association, speaking to each member in the name of all the rest. The type of reader, then, defined the newspaper. Newspaper readership research followed a pattern parallel to the growing newspaper industry itself. As the newspaper industry became more sophisticated and more competitive in the United States, publishers realized the implications of Tocquevilles writings about like-minded readers. Publishers gradually became more interested in putting faces and personalities on the anonymous recipients of their messages. However, newspaper readership also fit neatly into Americas post-World War II obsession with demographic research. A series of research reports by the Newspaper Advertising Bureau and others probed extensively into the characteristics of regular newspaper readers and the percentage of the populace reading newspapers or sections of newspapers at any one time (Bogart 1981).

30 By the 1960s, the Newspaper Advertising Bureau was tracking the number of American newspaper readers and their content preferences on a regular basis (Newsweek 2000). Part of that task was picked up by the research wing of the Newspaper Association of America when it absorbed the NAB in 1992. Like the NAB studies, the NAA studies point to

essentially the same predictors: higher age, higher education, and higher income equal higher newspaper readership (Stone 1987). The work of tallying readers was also taken on by independent research consultants such as Scarborough and Belden Associates. A similar mix of newspaper marketing organizations and private research firms exists in Canada and most other western countries. As Beam (1995) pointed out, the increasingly common use of research by newspapers is a reflection of a "marketing concept" that urges publishers to tailor their product to the informational wants and needs of their customers. The marketing concept is a familiar term in business literature, dating back at least 200 years to Adam Smiths statement that consumption is the sole end and purpose of all production; and the interest of the producer ought to be attended to only as far as it is necessary for promoting that of the consumer (Tull and Kahle 1990, Page 7). Various modern business scholars have defined it as a corporate state of mind, the external consumer orientation, and the practice of an organization aiming all its efforts at satisfying its customers at a profit (Houston

31 1986 Pages 82-83). In sum, it is synonymous with having a customer orientation. The media business has two sets of customers advertisers and the audience (Redmond and Trager 1998). Therefore, any media company that attempts to implement the marketing concept must address the needs of the audience. Beam (1996; 1995) said the marketing concept and its affiliated research has had a profound effect on U.S. daily newspapers by shifting the responsibility of defining content from the professional judgment of journalists to the desires of readers. As newspapers struggled with

declining numbers of subscribers, they enlisted market researchers to learn what readers wanted, then attempted to tailor the content of the newspapers to meet those desires. Beam juxtaposed this trend of administrative research with complaints from editors that the implementation of the marketing concept was reducing their control over news content and corrupting their professional standards. In his study of 360 newspaper editors, he found that as the uncertainty about a newspapers environment the system of customers, employees, management and government in which it operates increased, the more likely publishers were to use readership research and to implement marketing tactics. Using a fictional newspaper, he described the process:

32 Traditionally, the Daily Record" has relied upon professional journalistic judgment to guide content decision making. The Records newsroom may have established a structure for reporting about its community that was based on what its journalists believed readers needed to know. Their judgments in this regard may have resulted in an acceptable fit between the Daily Record and its environment. As the environment changed, however, uncertainty about the goodness of fit increased. In response, the Daily Records managers changed strategy to emphasize ascertaining what Daily Record readers said they wanted or needed in their newspaper and providing those wants and needs. This new strategy de-emphasized relying strictly on the professional judgments of Daily Record journalists to make content decisions. Thus, a change in the environment heightened perceived uncertainty, which in turn produced the shift toward a stronger marketing orientation (Beam 1996, p. 295). While this administrative focus on the newspaper audience has, in some aspects, empowered the reader, it is not without critics. Media scholar Jim Willis (Willis 1998) warns that the natural consequence of this process may be a metamorphosis from mass media to class media. Willis noted that as newspapers chase subscribers, they begin to pander to apathetic readers the so-called marginal readers that marketing experts identify as top prospects for new subscriptions. But newspaper publishers cannot afford to cater to every readers whim, Willis said, so they may be better off tailoring their product to fewer-in-number, but more appreciative, higher-income readers who are also good prospects for advertisers. Reuven Frank, the former president of NBC News, expressed the concerns of journalists in a more colorful way:

33 This business of giving people what they want is a dope pushers argument. News is something people dont know theyre interested in until they hear about it. The job of a journalist is to take whats important and make it interesting. (Hickey 1998, p. 34)

Reading Frequency and Media Penetration

The rate of media use is a central issue to all newspaper managers because it is the key factor for success or failure. For a newspaper, media use by the public is readership. Likewise, media use by the business community is advertising. The basic premise for newspaper operation in the United States today is that greater readership will beget greater advertising, as business owners will use the medium that puts their message before the greatest number of potential customers. That is, however, somewhat simplistic. Certainly price of advertising and quality of editorial product are factors in the success of a newspaper. More important for this study, however, are the concepts of potential audience and penetration. A newspaper, like any other consumer product, is produced not just for those who regularly use it, but for all those in the geographic or demographic market who are capable of using it. This is the audience. Success at reaching that audience is measured in terms of penetration. Newspaper penetration is commonly stated in terms of the

34 percentage of all households or households of a certain description within a geographic market that receive the newspaper. Although newspapers and other media are increasingly turning to "targeted markets" or specific demographic groups within a geographic area, the importance of overall penetration cannot be underestimated. High penetration makes a newspaper an efficient means of distributing advertising in a community while low penetration invites competition from other daily newspapers, weekly newspapers, catalogs, television and direct mail. As a Forbes analyst said: What makes household penetration critical is that declining penetration makes newspapers less attractive to advertisers over time. Their monopoly still exists, but it is increasingly less important and therefore less valuable as advertisers turn to other media to communicate their message. (Chakravarty and Torcellini 1989, p. 83) The previously mentioned drop in newspaper circulation as a portion of the total U.S. population the medium's penetration is, then, a concern for newspaper publishers and has supplied fodder for research. One older hypothesis for the decline in newspaper readership as a percentage of population was that lingering illiteracy, conflicting demands for time, and competition from television have simply driven Americans from the written medium (Sohn 1975). There is little support for that theory, however, when one looks at readership trends in other media. Circulation of nationally audited magazines, for instance, grew

35 from 189 million in 1960 to 367 million in 1998 (MPA 1999) and the household penetration of the nation's weekly magazines, both paid and free, went from 43 percent in 1970 to 56 percent in 1988 (Miller 1989, p. 7). Receiving a newspaper, however, does not necessarily mean using it as the editors and publishers intended. For the newspaper's message both editorial and advertising to be successful, someone must read it. The American newspaper industry and the communications researchers who study it have had a longtime fascination with newspaper readers. For the industry, this fascination has a very practical basis. As Bogart said: The economic power of the press as the nation's number one advertising medium, the political power of the press at the national level, the civic power of the press in every community -- all depend on one thing: the loyalty of our readers. (Bogart 1988, p. 1) One frequent measure of that loyalty is "frequency," or the number of times each week a person reads a newspaper. Studies that attempt to determine what makes the regular newspaper reader different from the infrequent or non-reader are scarce, however. The landmark early research was Westley and Severin's study of Wisconsin residents. That study provided the first clear picture of the non-reader as low in income, education and occupation scale; either quite young or old and likely to be a rural resident. The non-reader was also

36 found to belong to few formal organizations and have no political leaning (Westley and Severin 1964). Overall, however, they found that

newspapers were read by the vast majority of Americans: The single generalization that seems to lurk in these data might go something like this: the newspaper reaches nearly everyone except those who tend to be relatively isolated, both by distance from neighbors and by a relative lack of social contact with others, in both formal and informal settings. (Westley and Severin 1964, p. 49) Westley and Severin also found no difference in the use of other media by readers and non-readers of newspapers. Rarick (1973) took a different tack when he attempted to distinguish between daily newspaper subscribers and non-subscribers. Although he was careful to point out that a subscriber is not necessarily a reader, and vice versa, he did find that subscribers tended to come from households in which the "breadwinner" had a higher income than in the typical non-subscriber households. Similarly, the chief breadwinner in a subscribing household tended to be better educated and between 35 and 64 years old. These findings seemed to support Westley and Severin. The Westley and Severin study was partially replicated in 1974 by Penrose, Weaver, Cole and Shaw. While they found that the same type of person who was not reading a daily newspaper 10 years earlier was still not reading one, there were many more nonreaders than Westley and

37 Severin found. The researchers attributed this primarily to a decline in the educational level of the populations surveyed by the two studies. Rather than seeking to move more people from the non-reader to the reader group, Penrose, Weaver, Cole and Shaw recommended the newspaper editor increase the amount of in-depth material: ... for the bird he clearly has in hand ... the more highly educated, urbane audience of the middle age range. In other words, to overly democratize the news presentation is to provide material that is not truly useful or interesting to the most important audience of newspapers. (Penrose et al. 1974, p. 636) Poindexter (1979) moved beyond the demographic statistics of earlier studies by asking survey participants why they did not read daily newspapers. The most common reasons given were lack of time, use of another news medium, cost and lack of interest in the content. Note that the use of another news medium as a reason not to read a newspaper conflicts with the earlier findings of Westley and Severin. Poindexter found two distinct subgroups of non-readers. Her "typical non-readers" are at the extremes of the age continuum (very young or very old), poor and undereducated, as other researchers had noted. However, she found a group of non-readers who fell outside these bounds. Her "atypical non-readers" have essentially the same demographics as readers but say they do not read because of lack of time and newspaper content.

38 In attempt to further refine readership numbers, Gollin and Salisbury (Gollin and Salisbury 1980) studied three analytical methods used in readership demographic studies. Using data collected during a 1977 survey for the Newspaper Advertising Bureau, they attempted to determine what demographic variables differentiated people who read a paper four or more days a week (frequent readers) from those who read less than four days a week. The same data were analyzed via multiple regression, the Automatic Interaction Detector program and cross-tabular analysis. All three methods identified income, education, geographic region and age as key demographic factors, although cross-tabular analysis also identified marital status, race and length of residence as key factors. Frequent readers, Gollin and Salisbury said, are apt to be more than 35 years old, well-educated, affluent, white and live in the Northeast or North Central regions of the United States, and have lived in their current home for more than five years. However, Gollin and Salisbury noted that the multiple correlation of the demographic variables was relatively low (r=.398) and therefore concluded that demographics provide limited predictive power for newspaper readership researchers. Sobal and Jackson-Beeck (1981) used a stricter definition of "nonreader" than earlier studies, counting only those survey respondents who answered "never" to a direct question asking how often they read a

39 newspaper. Consequently, they found that a much smaller proportion of the general population fell into the non-reader category than did other studies. Sobal and Jackson-Beeck's survey identified 4.5% of their national sample as non-readers of daily newspapers. Westley and Severin found 13.5% of respondents in Wisconsin said they did not generally read a newspaper. Penrose, Weaver, Cole and Shaw, found 31% of their North Carolina respondents had not read a newspaper today or yesterday. Poindexter found 10% of her New York sample said they seldom or never read a daily newspaper (Sobal and Jackson-Beeck 1981, pp. 9-10). Demographically, Sobal and Jackson-Beeck's chronic non-readers are significantly older than readers, much less educated, more likely to be a part of an ethnic minority, poorer, likely to live in open country and not in suburbs, and more numerous in the South Atlantic and South Central states. They were more likely than readers to be liberal and affiliated with the Democratic Party, but less likely to vote. Significant in the Sobal and Jackson-Beeke study is the finding that chronic non-readers represent an extremely small percentage of the population as a whole. Yet, as cited earlier, declining newspaper penetration is a major problem for the industry. Increasingly, the focus of the readership conundrum is on the infrequent reader. Infrequent newspaper readers defined as those who read a paper one to three days of the past five weekdays made up 28% of the sample

40 used in a Newspaper Advertising Bureau study in 1985. They tended to be younger, have slightly less income and rent their home. They were likely to be single or divorced/separated/widowed. However, they were evenly divided among educational levels and size of the city or town they live in (NAB 1978). The NAB study also found frequency of reading strongly correlated with getting home delivery of the paper, alluding to the power of the reading habit and suggesting newspapers find ways to concentrate on the needs of these infrequent readers to boost circulation figures. Frequency of readership continues to be an issue in newspaper marketing. The Newspaper Association of America's on-line sales tool, "Why Newspapers?," includes a section entitled "A Habit Forming Medium." In that section, the NAA reports a 1997 study that found that more than half (51%) of American adults are frequent readers of newspapers, meaning they read four or five weekday editions a week. The figure is higher for men (53%) than women (49%) and increases as both education and income go up (NAA 1998). A related NAA document showed that a sense of "rootedness" in one's community is related to reading weekday papers. People who have lived in the area more than 10 years, the NAA said, are nearly twice as likely to be regular newspaper readers as are newcomers (NAA 1998, p. 15).

41 Studies on the frequency of newspaper reading, while helpful to media managers striving to retain circulation revenue, also raised questions of volition and the effect of influence that could only be answered by theoretical research.

Theoretical Research

While demographic researchers have counted the number of readers and their preferences, a parallel stream of research sought to find clues to why people read a newspaper or consume other media. That question also produced a philosophical question in reader research: Do people read for personal reasons, or do people read for societal reasons? The introduction of television in the 1950s and 1960s prompted media researchers to focus their efforts on short-term effects of mass media exposure. Those studies, however, failed to find significant linear effects, which in turn prompted some researchers to look at the interests audience members brought with them when they turned on the television or picked up the paper (McQuail 1994). A body of research focusing on a dual perspective on media effects media output and audience came together as the uses and gratifications model, usually identified with the 1974 book The Uses of Mass Communications edited by Blumler and Katz.

42 Drawing on the traditions of psychology, the uses and gratification school asserted that consumption of media fulfilled basic human needs (Blumler and Katz 1974). A series of studies in the 1940s and 1950s developed lists of functions served either by specific media content or by a medium in general. These functions of media included: to match ones wits against others, to get information for daily living, to provide a framework for ones day, to get the cultural tools to become upwardly mobile, and to be reassured about ones own usefulness (Katz, Blumler, and Gurevitch 1974). Uses and gratification theory draws strength from Maslow's hierarchy of needs, a staple of psychological literature that outlines the personal needs of the individual (Rosengren 1974). Uses and gratification will be explained in greater detail in Chapter 3, but suffice it to say that the model developed a large body of literature linking the individual's needs with the consumption of mass media. While the uses and gratifications school focused on the individual, another group of scholars was looking at the society. A study by Chaffee and McLeod (1973, p. 237) noted that use of mass communications occurs not in isolation from the rest of a person's social life, but rather is interwoven in "an outgoing system of reciprocal influences." The study showed that social utility plays a major role in the type of information people select.

43 Scholars steeped in the sociological tradition addressed the same question explored by uses and gratifications why do people use the media from the broader societal perspective. From this research developed media systems dependency theory, which also will be discussed in detail in Chapter 3. The theory or more precisely, the model postulates that people turn to the media for social and emotional support when urbanization, industrialization or modernation distances them from the comforting mantle of family, church and neighborhood. Loges and BallRokeach showed that readers who demonstrate high levels of such dependency also spend more time with their newspapers (Loges and BallRokeach 1993).

Convergence

The field of audience studies showed strong signs of theoretical convergence in the last decade of the 20th century. Administrative and theoretical research goals are often blended, as in Stone's study of the habits of readers whose newspaper switched from evening to morning circulation (Stone and Stone 1990). Researchers have also taken steps toward determining the driving motivations for reading of all kinds, not just newspaper reading. The International Adult Literacy Survey in Canada, for instance, documents the role of "fun" in daily reading habits (Crompton 1997).

44 The American Society of Newspaper Editors, the Newspaper Association of America and the Media Management Center at Northwestern University pooled resources in April 1999 to form the Readership Initiative a prime example of administrative and theoretical convergence. The five-year project has many of the same circulationbuilding goals of traditional administrative studies, but focuses on determining what drives people to read the paper rather than on whether they are reading a particular section (Nesbitt 2000). The first phase of the initiative was launched in January 2000 when a stratified sample of 104 newspapers was selected for a multidimensional study. The study will combine a reader survey with a two-year qualitative analysis of the papers and their markets to determine how all three combine to drive readership. One Oregon newspaper, the

Corvallis Gazette-Times, was selected to participate in the project. The initiative is just the latest in a long line of newspaper audience studies set in motion by a perceived industry problem (such a declining readership) or societal challenges (such a adult illiteracy). Nevertheless, audience research has been an productive platform for theory building. Chapter 4 will discuss this aspect of the field.

45

CHAPTER III

READERSHIP THEORY

Early theoretical discussions of newspaper readership focused on its role in citizenship. Both Alexis de Tocqueville (Tocqueville 1969) and Thomas Jefferson (Mott 1962) remarked on the relationship between the voting public and the mass media primarily newspapers at the time and the role of newspapers as the "eyes" of a democracy. Lippmann (1922), however, took some of the first steps toward a true theory of newspaper readership when he eloquently explored the creation and dissemination of public opinion. Lippmann was arguably one of the most influential journalists of the early 20th. His intellectual but easy-to-follow style was often imitated, but usually poorly eventually giving way to the hard-edge journalistic style now popular. Nevertheless, the quality of his statements earned him a page and a half of text in Bartletts Familiar Quotations (Bartlett 1968 (1882), pp. 1012-1014). Of Lippmanns many books, Public Opinion is often quoted as the seminal discussion of why Americans think as they do. Godfrey Hodgson, director of the Reuter Foundation Programme at Oxford University, calls it one of the few undisputed classics on the long but not overdistinguished shelf of writings about journalism.

46 Public Opinion is probably best remembered for Lippmann's treatise on the world outside and the pictures in our heads" (Lippmann 1922, p. 18). The essay was both a discussion of humankind's inability to perfectly perceive the world and a description of how people use the mass media. None of us, he explained, can see the whole world as it is. We take shortcuts via media use, censorship, contact with others, values and stereotypes to form pictures of the world as we imagine it might be. But these shortcuts, he said, represent significant challenges to the concept of mass communications. The problem created for the mass media is twofold. Journalists

themselves have these shortcomings of perception when they report the news. Then the readers or listeners filter the message of the journalists through their own set of mental pictures. As Lippmann said: Each of us lives and works on a small part of the earths surface, moves in a small circle, and of these acquaintances knows only a few intimately. Of any public event that has wide effects we see at best only a phase and an aspect. This is as true of the eminent insiders who draft treaties, make laws, issue orders, as it is of those who have treaties framed for them, laws promulgated to them, orders given at them. Inevitably our opinions cover a bigger space, a longer reach of time, a greater number of things than we can directly observe. They have, therefore, to be pieced together out of what others have reported and what we can imagine. (Lippmann 1922, p. 53) Although he only brushed on it tangentially, Lippmann was wrestling with a concept that came to define audience theory: How active

47 is the human mind in the selection and interpretation of mass media information? Do humans pick and choose the messages they want to remember or take action on, or do they let mass media message wash over them like unchecked waves sometimes battering, sometimes washing and sometimes refreshing them? Biocca notes that the dichotomy of perspectives on audiences has come to permeate theoretical debate in mass communications research. On one end of the rope we find the active audience: individualistic, impervious to influence, rational and selective. On the other end, we have the passive audience: conformist, gullible, anomic, vulnerable, victims. Hugging and tugging at each end is an assorted lot of key media theorists championing their perception of social reality. (Biocca, p. 51) The use of propaganda by both sides in World War I brought about a keen interest in the effect of mass media messages on the public. After the war, the discussion by Harold Lasswell, Leonard W. Doob and others about the omnipotent power of the media to change minds came to be known as the "magic bullet" or "hypodermic needle" theory (DeFleur and BallRokeach 1989). The theory postulates that audiences are passive and allow the messages of the media drive deep into their psyches with little or no resistance. The resulting effect can be direct and immediate or it can be an indirect effect over time (and as the result of exposure to repeated messages). The passive audience concept was and still is of great interest

48 to social commentators concerned with the ownership and control of media organizations. The notion of a passive audience, Biocca said, also continues to flourish where media discussions turn to the fear of the rise of mass society, such as in cultivation theory (Biocca ). Rouner alluded to this in her discussion of "non-selective" and "habitual" television viewing habits (Rouner 1984, p. 168). Between the two world wars, most of the research and theory development employing the magic bullet theory was more clearly in the media effects arena than in audience studies (Nowak 1997). But in the search for universal effects of media messages, scholars found puzzling evidence that some audience members paid attention to a message while others ignored it. This implies that audiences are "active" they choose which messages to process. It is this concept of active choice and therefore questions about what influences the choice and whether different people choose messages in different ways that became the basis for most modern audience scholarship (Biocca 1988; Katz, Blumler, and Gurevitch 1974; Livingstone 1998). Much of the theoretical support for the active audience concept is based on the writing of social psychologist Raymond Bauer in the 1960s. Bauer lashed out against the one-way influence model of psychologist B.F. Skinner and his followers. He favored instead a transactional model where

49 the reader or viewer insists that exchanges of information are equitable. His model is sometimes called the "obstinate audience" (Bauer 1963). Blumler identified four characteristics that can be used to measure audience activity: utility, intentionality, selectivity and imperviousness

to influence (Blumler 1979). Levy and Windahl examined Blumler's contention empirically and demonstrated that audience activity is not only an important dimension of media use but that activity also varies between audience members. The citation of social psychologist Bauer for a communication theory as audience study developed was not unique. As communications studies was a relatively new field of social science, early researchers drew on the theorists of other fields to support their hypotheses. This led to a theoretical split between audience researchers who worked from psychology's focus on the individual and those who worked from sociology's focus on the group. From the psychological perspective grew the uses and gratifications model. The sociological perspective produced the media system dependency model.

Uses and Gratifications

Early proponents of the uses and gratifications model had difficulty explaining the theory upon which it is based it was, in fact, often linked to work that seemed to disclaim theoretical pretensions (Katz, Blumler,

50 and Gurevitch 1974, p. 21). In the late 1950s and early 1960s, the introduction of television into Western culture sparked considerable interest in the impact of mass media especially upon children. Media studies had a broad scope at the time, but concentrated on effects and functions (McQuail 1998). In 1959, Elhu Katz called for a new perspective in media studies. He said that less attention should be paid to what the media do to people "and more to what people do with the media" (Katz 1959, p. 2). The approach, he wrote, assumes that no mass media content can have impact upon the individual who has no "use" for it. A variety of researchers took up Katz's call in the 1960s and 1970s for descriptions of what became known as the uses and gratifications model. Drawing on the traditions of psychology, the uses and gratifications school asserted that consumption of media fulfilled basic human needs. In 1974, Katz, Blumler and Gurevitch offered a now-wellknown clarification that defining the model with five "elements:" 1. The audience is conceived of as active. 2. In the mass communication process, much initiative in linking gratification and media choice lies with the audience member, limiting straight-line media effects. 3. The media compete with other sources of need satisfaction. 4. Methodologically speaking, many of the goals of mass media use can be derived from data supplied by individual audience members themselves. 5. Value judgments about the cultural significance of mass communication should be suspended while audience

51 orientations are explored on their own terms. (Katz, Blumler, and Gurevitch 1974, pp. 20-21) As uses and gratifications developed, some theorists challenged the absolutist tenor of the list conceived by Katz, Blumler and Gurevitch. Rubin eventually consolidated the literature to fine-tune this list into a set of five assumptions based on the premises that media audiences are variably active and that researchers must understand audience behavior and motivations if they are to explain media effects: 1. Communication behavior such as media use is typically goal-directed or motivated. Such behavior is functional for people: it has consequences for people and societies. 2. People select and use communication sources and messages to satisfy felt needs or desires. Media use is a means to satisfy wants of interests such as seeking information to reduce uncertainty or to solve personal dilemmas. 3. Social and psychological factors mediate communication behavior. Behavior is a response to media only as filtered through one's social and psychological circumstances such as the potential for interpersonal interaction, social categories and personality. 4. Media compete with other forms of communication for selection, attention and use. There are definite relationships between media and interpersonal communication for satisfying needs or wants. 5. People are usually more influential than media in mediaperson relationships. (Rubin 1993, p. 98) A notable point in Rubin's explanation of the uses and gratifications tradition is that, despite the inclusion of social factors into his five assumptions, he pointedly declares that the assumptions illustrate that uses and gratifications is a psychological communications perspective

52 because the uses are controlled by the active receivers of media information (Rubin 1993). A hallmark of uses and gratifications research in is the definition of typologies of media users. These typologies identify areas of "use" for which people employ mass media to "gratify." A series of studies in the 1940s and 1950s had developed lists of functions served either by specific media content or by a medium in general. The lists of functions developed into typologies of readers who could be linked to the functions. These functions of media included to match ones wits against others, to get information for daily living, to provide a framework for ones day, to get the cultural tools to become upwardly mobile, and to be reassured about ones own usefulness (Katz, Blumler, and Gurevitch 1974). The typologies varied from researcher to researcher, but often were similar to Maslow's hiearchy of needs (Rosengren 1974). A psychology theorist, Maslow said all people strive to satisfy five needs in the following order of importance: 1. 2. 3. 4. 5. Physiological needs Safety needs Belongingness or love needs Esteem needs A need for self-actualization (Maslow 1954)

Lasswell outlined one of the first typologies: surveillance, correlation, entertainment and cultural transmission (Lasswell 1948).

53 Berelson focused on what readers themselves reported as their "special" uses for newspapers: information and interpretation of public affairs, tools for daily living, respite, social prestige and social contact (Berelson 1949). McQuail and his associates modified earlier lists to surveillance, personal relationships, diversion and personal identity (McQuail, Blumler, and Brown 1972). As what came to be known as uses and gratifications research became more sophisticated, the typologies became more specific and complex. By the late 1970s, uses and gratifications research had developed a reputation as "functionalist in conception, individualist in method of collection and lending itself to multivariate analysis" (McQuail 1998, p. 153). This opened it to attack by the critical studies and qualitative methodology camps of communications scholarship. Bybee (1987) took issue with the gratifications approach, particularly in the way it regarded the nature and extent of audience activity. Re-reading the literature from a political communication perspective, he also took issue with the way uses and gratifications researchers assumed the audience had the ability to consciously recognize and act in its own best interest. In addition, he objected to the limited conceptualization of the distribution and maintenance of political power. Although uses and gratifications research continues, the popularity of the tradition declined in last two decades of the 20th century.

54

Media System Dependency

In many ways, the media system dependency theory grew out of a sense of frustration with uses and gratifications by communications researchers who subscribed to the sociological school of human motivation. Blumler and Gurevitch demonstrate this frustration from within the uses and gratifications school when they bemoan the paucity of attention communication scholars give to the processes of communication in favor of the study of mass media structure. They attribute this focus to the two dominant paradigms of the postwar period: The scientific model of inquiry that emphasizes objective observation and is biased by the assumption that society is more or less stable. The anti-positivist movements of cultural studies, critical theory and hermeneutics that argue the media impede significant change rather than promote it. (Blumler and Gurevitch 1996) The researchers argued that the social system and the mass media are interdependent. Viewed vertically, the mass media are mediating agencies between sources and the audience. Viewed horizontally, they provide diverse channels through which groups disseminate their claims on resources, status, identity and power.

55 Ball-Rokeach, recalling the intellectual process that led her to become one of the founders of media dependency theory, said she had much in common with the pioneers of uses and gratifications but also had a significant disagreement with them on the nature of the audience. I was as dissatisfied with mechanical models of media effects and their conception of passive audience members as were Katz, Blumler, and others who were developing the U&G perspective. However, the move to an active individual who employs interpretive powers to override the influence of creators of media texts was not the kind of active individual I saw around me. (Ball-Rokeach 1998, p. 9) Ball-Rokeach and co-author Melvin DeFleur launched the media systems dependency school of audience research with "A Dependency Model of Mass Media Effects" in 1976 (Ball-Rokeach and DeFleur 1976). She later wrote that she could not visualize an audience driven by firm psychological imperatives. Instead, she found an audience wracked with ambiguity due to conditions of structural alienation and conflict conditions over which they had little control (Ball-Rokeach 1998). She

postulated that people turn to the media for social and emotional support when urbanization or social factors distance them from the comforting mantle of family, church and neighborhood. Rather than gratifying a basic need, the media then become social support surrogates upon which the reader comes to depend. Ball-Rokeach defines media system dependency as a relationship in which the capacity of individuals to attain their goals is contingent upon

56 the information resources of the media system (Ball-Rokeach 1985, p. 487). Those resources are the systems capacity to gather, process and disseminate information. The model refers to goals instead of needs to connote a problem-solving motivation in which individuals may not consciously articulate their dependency relationship with the media, yet have the ability to articulate goals that give rise to media behavior. Loges and Ball-Rokeach showed that readers who demonstrate high levels of such dependency also spend more time with their newspapers (Loges and Ball-Rokeach 1993, p. 30). Rubin found a similar sense of heightened media dependency when physical barriers restrict social interaction. His studies showed that less healthy and less mobile people depend more upon television than do self-reliant people (Rubin 1993). If the media and social systems are interrelated, changes in one affect the other. Changes in the media have included a concentration of ownership and related financial structure, the expansion of the media market through the development of new technology, the internationalization of media outlets, improvements in the technology of dissemination and audience fragmentation. This sense of "relationship" was fundamental to the argument BallRokeach and DeFleur first outlined in their "dependency model of mass media effects" in 1976. Ball-Rokeach and DeFleur said that mass

57 communication is a function of complex relationships between the media, audiences and society (Ball-Rokeach and DeFleur 1976). By adding "system" to the media dependency term, the researchers showed that these relationships were much greater than a single session between a viewer and a television program or a reader and a newspaper. In common usage, though, the terms media system dependency and media dependency are interchangeable (Merskin 1999). The dependency model looks at how audience members rely upon the broad concept of mass media to provide them with information, advice and services they might have found elsewhere in a less complex social structure. Ball-Rokeach defines media system dependency as a relationship in which the capacity of individuals to attain their goals is contingent upon the information resources of the media system (Ball-Rokeach 1985, p. 487). Those resources are the systems capacity to gather, process and disseminate information. The model refers to goals instead of needs to connote a problem-solving motivation in which individuals may not consciously articulate their dependency relationship with the media, yet have the ability to articulate goals that give rise to media behavior. Merskin (1999) said that the concept of mass society upon which dependency theory hangs makes three assumptions: 1. Individuals are in a situation of psychological isolation from others. 2. Impersonality prevails in social interactions.

58 3. Individuals are relatively free from the demand of binding obligations. (Merskin 1999, p. 116) As social systems become more urbanized and industrialized, people tend to rely more and more upon the mass media in their coping strategies, instead of more traditional support systems such as family, friends, church and school. Ball-Rokeach breaks the media dependencies of individuals into three functional categories: Understanding (both social and self), orientation (both action and interaction), and play (both social and solitary). In these three areas, the use of the media as a coping guide (or to stay in touch with the world) in modern societies has become so predominant that, as Ball-Rokeach said, There are few, if any, functional alternatives to the media system for the average American. (BallRokeach 1985, p. 494-496) Interpersonal networks are still important in social discourse, but the media both shape and serve interpersonal discourse. Such discussion is often centered on issues and events framed by the media. Just as uses and gratifications has its five "elements," the media dependency model can be viewed through five "propositions" listed by DeFleur and Dennis (1996)and reiterated by Merskin (1999): 1. People in all societies need information in order to make numerous decisions about political affairs, obtaining food and shelter, transportation, mating, and many other aspects of daily life.

59 2. In traditional societies, people tend to pursue similar ways of life, and are linked to word-of-mouth networks of extended families, long-term neighbors, deeply established friendships and other channels from which they obtain the information they need. 3. In urban-industrial societies, populations are composed of unlike people brought together through internal migrations and immigrations from outside. They are greatly differentiated by such factors as race, ethnicity, occupational specialization and complex economic classes. 4. Because of their far greater social differentiation, people in urban-industrial societies have fewer effective word-ofmouth channels based on deeply established networks of social ties through which they can obtain the information they need in daily life. 5. People in urban-industrial societies depend heavily on mass communications for information needed to make many kinds of decisions. From the media they obtain a flow of information, advice and role models contained in the news, entertainment, and advertising which they use as a basis for those decisions. (DeFleur and Dennis 1996; Merskin 1999) Theorists assert that people can develop dependencies not just on media systems, but also on a specific medium, such as television or a newspaper. Various media, from television to newspaper classified ads, have been shown to be crucial to the purchase decisions of consumers. (Becker and Whitney 1980; Mueller and Kamerer 1995; Gaziano 1990) Media dependency theory offers scholars a way of exploring changes within modern society while taking advantage of the interdisciplinary nature of the field. Intellectually, media dependency provides a way of organizing human social behavior within a model that suggests a logical flow of information between senders and receivers and a return flow of

60 support and loyalty from receivers to senders. It also posits that in the media people find not only information, but also society-building advice and role models.

Comparison and Convergence

Both uses and gratifications theory and media dependency theory have come under fire for their limitations. Dependency only explains why people rely on mass media, while uses and gratifications only focuses on what the consumer does with the media (Baukus 1999). Uses and gratifications is seen by some scholars as overly voluntaristic immune from societal or media pressures. Dependency theory, on the other hand, is seen by some as too deterministic under-attributing the power of individual will (Rubin and Windahl 1986). Ball-Rokeach addressed the criticism with an exhaustive comparison of the two theories of media audience studies. Her findings are summarized in Table 1. She found that the two theories shared a view that the audience member is active, but differed fundamentally in how that activity is framed. Uses and gratification theorists, she wrote, say audience members use the media to gratify internal psychological needs. Media dependency theorists, she said, say audience members use the media as sociological tools to solve problems in their daily lives (BallRokeach 1998, pp. 27-28).

61 TABLE 1. Ball-Rokeach's Comparison of Uses and Gratifications with Media Systems


Comparative Topic
Theoretical origins Experimental origins Central question

Uses and Gratifications


Perception, attitude diffusion theories The life and issues of the 1940s and the 1950s How do individuals, alone or in small communities, reconstruct media texts to gratify their needs? Selectivity Individual and network Individual in interpersonal environs Interpersonal sources of variation in uses Major sources of variations in use

Media Systems Dependency


Power-dependency theory and social ecology theory The life and issues of the 1960s and the 1970s Why, when, and how are media powerful re: individuals & interpersonal networks & with what consequences? MSD relation Individual, network and macro Interpersonal networks in ecology of macro MSD relations Mass effects interpersonal text creation & interpersonal affects procession of mass texts Minor source of variation in characteristics of MSD relations

Central concept Analysis units Locus: knowledge construction Interpersonal vis-a-vis mass communication Individual differences Conception of: Audience members

Active text interpreter grounded in interpersonal environs and unconstrained by macro production forces Text creators NA NA Selective by individual needs Cointerpreter and buffers of media influence

Active text interpreter grounded in personal, interpersonal & social environs, constrained by macro MSD relations of production Information system central to conduct of personal & social life Control of dependency-engendering information resource Determined by ecology of macro MSD relations Selective by MSD relations Hold MSD relations that affect individual MSD relations

Media systems Media power Production Consumption Interpersonal networks Method of: Observation Analysis Interpretation

Ask people about uses a la empirical factors Factor analysis and regression; reception analysis Individual differences situated in demographic and interpersonal

Ask people about MSD relations a la conceptual technology. Confirmatory factor analysis and regression MSD relation differences situated in personal-social environs and in micro MSD relations.

From Ball-Rokeach, Sandra J. 1998. A Theory of Media Power and a Theory of Media Use: Different Stories, Questions and Ways of Thinking. Mass Communication and Society 1 (1-2):5(40).

62 The criticism that uses and gratifications is too individualistic and psychological also spawned another branch of research, primarily in Europe. Reception analysis focuses on the encoding of messages and the types or genres of messages that make multiple interpretations of text possible. A variant, media ethnography, focuses on contexts (Reimer 1997). However, as both these qualitative research fields concentrate more on media content then on media use, they are not applicable for this dissertation. Nevertheless, the enthusiastic entrance of qualitative researchers into the audience studies field has had an impact on both uses and gratifications and media dependency. By the mid 1980s, traditional audience researchers were citing the arguments of reception analysis and media ethnography and vice versa. In 1987, Schrder wrote: The work of prominent scholars in the sociological and critical paradigm appears indeed to be converging even to the point where some of the projects become indistinguishable with respect to underlying theories and hypothesis, research design, analytical method and (in general terms) substantial results. (Schrder 1987, p. 28) Reimer argued that audience studies at the end of the 20th century were too focused on micro issues of personal media use and the context of particular text. He called for a more eclectic form of research that looked at both micro and macroscopic issues in media use (Reimer 1997).

63 Uses and Dependency

Rubin and Windahl put flesh on the ethereal notion of theory convergence with the "uses and dependency model " (Rubin and Windahl 1986). They argued that, taken alone, dependency is too rigid because it traces all media use back to social-structural influences. Uses and gratifications, on the other hand, leaves too much to choice and grants to audience members the ability to select media elements to make sense of the content. Their solution was to combine the two theories into a single, flexible model. The interaction of audiences and media is more complicated than either dependency or uses and gratifications imply, according to the uses and dependency model. Rather than assuming that human needs are basic and stable, Rubin and Windahl said, "We propose that people's needs and motives vary as they evolve in interactions with societal and communications systems" (Rubin and Windahl 1986, p. 186). The uses and dependency model envisions a highly interactive system in which the audience member is influenced by both the societal system and the mass media system, uses both the mass media and functional alternatives and is responsive to the effects or consequences of thoughts or behavior (Figure 2).

64

SOCIETAL SYSTEM Sociocultural Structure Political Structure Economic Structure

MASS MEDIA SYSTEM Content Structure Functions

AUDIENCE Psychological Traits Social Categories Social Relations Needs Interests Motives

FUNCTIONAL ALTERNATIVE USE Non-Media Channels Media Channels and Content Consumption Processing Other Activity Dependency Non-Dependency

MASS MEDIA USE Mass Medium Media Content Consumption Processing Interpreting Dependency Non-Dependency

EFFECTS OR CONSEQUENCES Cognitive Affective Behavioral

From: Rubin, Alan M., & Sven Windahl. 1986. The Uses and Dependency Model of Mass Communication. Critical Studies in Mass Communication 3:184-199.

FIGURE 2. Uses and dependency model

65 Media dependency in the system is situational. As Baukus explained: Within a given set of socially constrained media alternatives, it is up to the person to evaluate the potential of each medium to fulfill personal information requirements. (Baukus 1999, p. 96) A key characteristic of the uses and dependency model is the manner in which it describes the "outcomes" of media use. Rubin and Windahl assert that media use results in both effects and consequences. Effects are the result of active, "instrumental" use of media content. An example might be a reader's decision to view the 8:30 p.m. showing of "Titanic" after looking up the movie listings in the newspaper. Consequences, on the other hand, are simply the result of the actual use of media and therefore exposure to the content. Consequences are the "outcome of ritualized use of a medium" (Rubin and Windahl 1986, p. 196). An example might be humming the tune to the "Jeopardy" game show as your son tries to decide what ice cream to buy. This type of media use may engender a feeling of belonging, or it may displace other, nonmedia activities in the audience member's life, the researchers said. Under the model, habitual (non-instrumental) media use may in and of itself lead to increased dependency on the very same medium.

66 To date, the uses and dependency model has yet to catch the imagination of the audience studies world. Only one dissertation employing the model is listed in Dissertation Abstracts (Taylor 1991). "Part of the problem is that the model requires both a micro and macro level analysis," said Donald Taylor, a professor at California State University at Sacramento and author of that dissertation. marry those two things methodologically?" (Taylor 2000) Despite the logistical problems which Taylor said also include the use of both quantitative and qualitative methods the model has great potential. "It really does offer a common sense way to understand media audiences because audience members are also members of a society, not from outer space or something" (Taylor 2000). For the purposes of this dissertation, the uses and dependency model offers a coherent framework to combine the concepts of active and passive audiences. The model posits that scholars from both the uses and gratifications and the media dependency schools are right, but that media use is not driven by a single force. Instead, it concedes the existence of a socially constrained set of media alternatives, but gives to individuals the power to evaluate the potential of a medium to fulfill their personal information requirements. The audience, then, is active but influenced. It is upon this convergent theory that this research will be based. "How do you

67

CHAPTER IV

HABIT AND RITUAL

Although only nominally studied in mass media communications, the power of habit and ritual has a long history of intellectual investigation in the annals of philosophy, sociology and psychology. The Greek philosopher Aristotle (384-322 BCE) wrote that habit which he defined as a quality of long duration and difficult to change can ease many of the discomforts of life. Concentration, hard study and intense effort are painful because they involve constraint and compulsion, he wrote. The pain melts away into pleasantness, Aristotle noted, when these burdens become habitual (Cooper 1932). Aristotle, Evenus, Publius and other thinkers of the Hellenic Age ascribed great positive power to habit. Hsieh argues that habit became the very backbone of Aristole's pursuit of "virtue," citing the many references to habit in the Greek's famous book on virtue, The Nicomachean Ethics (Hseih 1997). From that book comes one of Aristotle's oft-repeated quotes: "We are what we repeatedly do. Excellence then, is not an act, but a habit." Stone and Stone updated Aristotles theories with an analysis of another classic thinker St. Thomas Aquinas (Aquinas 1272 (1966)).

68 Aquinas postulated that habits are not just mindless acts, but powerful tools of human will. The Stones condensed the writings of Aquinas (12251274 AD) into five key arguments about habit: 1. Habits are more permanent than passive qualities. 2. A habit implies an ordering toward action. 3. Habits are an experience of the will, hence they are intentional. A habit can be made use of when a person wills to use it. 4. People differ in their habits. 5. With repeated acts, a habit grows, but a habit can also be diminished with the cessation of an act. (Stone and Stone 1990, Page 26) Philosopher and psychology pioneer William James waxed poetic about the power and social potential of habit at the beginning of the 20th century. James, the founder of the American pragmatism movement, drafted his thoughts on habit as part of his Principles of Psychology in 1890, but revised them into an entire book on habit in 1914. He said that rather than to be shunned, habit is to be encouraged among intellectuals (James 1914). His rationale was that habit reduces many actions into automatic responses that require no intellectual energy. The more routine we turn over to habit, the more our higher powers of mind will be set free for their own proper work. Habit, thus, is the enormous fly wheel of society, its most precious conservative agent. It alone is what keeps us in the bounds of ordinance...(James 1914, p. 51)

69 His contemporary, John Dewey, also described habit in glowing terms. Dewey called habits "arts" to be actively acquired: They involve skill of sensory and motor organs, cunning or craft, and objective materials. They assimilate objective energies, and eventuate in command of environment. They require order, discipline and manifest technique. They have a beginning, middle and end. (Dewey 1922, p. 15) At one time or another, the terms habit, rite, and ritual have been used interchangeably by theorists. Most tend to agree, however, that the selection of a term depends on the degree, intent and timing of the action. Although most scholars agree that rite is most properly used to refer to a specific ritual activity, the distinction between habit and ritual is less clear. Durkheim, in his 1915 work The Elementary Forms of Religious Life, established the view that rituals are the mechanism that produces ideas charged with social significance a concept that retains import in the journals of sociology today (Durkheim 1915; Collins 1985). Kluckhorn (1942)set a baseline for the study of ritual in his general theory of myths and rituals, published in 1942. Kluckhorn pored through the work of dozens of theorists and many case studies to arrive at a fairly simple definition: Ritual is an obsessive repetitive activity often a symbolic dramatization of the fundamental needs of the society, whether economic, biological, social, or sexual. (Kluckhorn 1942, p. 78)

70 In that context, routinely reaching for the newspaper to start my morning or settling into the recliner after work with the sports section with a nightly sigh of relief may be thought of as a type of ritual both social and biological. Wallace expanded on Kluckhorns definition by giving ritual a purpose: It averts misfortune or confers benefits to participants. In a sense, it provides a level of comfort (Wallace 1966). Habit has also been used to challenge popular notions of consumer behavior, of which newspaper readership is a natural subset. Kahle and Beatty (1987) tested a well-known theory of the attitude-behavior relationship. The theory of reasoned action, proposed by Martin Fishbein, assumes that consumer attitudes and behaviors are related and that behavior results from intention. Kahle's alternative theory of social adaptation (Kahle 1984) says that the presence of habit prohibits "reasoned action" from being as comprehensive as Fishbein and others claim. Kahle and Beatty designed a cross-lagged panel correlation quasi-experiment of coffee drinkers that on the face of the data supported Fishbein's theory. However, when the researchers used regression analysis to isolate the influence of habit, the effect of reasoned action diminished. A second study of soft drink users (Beatty and Kahle 1988) showed that habit was particularly powerful for consumers making low-involvement decisions.

71 The linkage to low involvement implies that habit or ritual can be a driving force in routine, day-to-day activities. Imber-Black and Roberts clarified the role of ritual in daily life by defining four types of common ritual: 1. Day-to-day essentials (eating, sleeping, greeting and parting) 2. Family traditions (anniversaries, birthdays, reunions, vacations) 3. Holiday celebrations (celebrations with family and friends but shared with the outside world) 4. Life-cycle rituals (baby showers, engagement parties, weddings and funerals) (Imber-Black and Roberts 1992, pp. 14-18) The habitual reading of the newspaper especially when the reader includes it in the "ceremony" of a daily meal such as breakfast fits neatly into Imber-Black and Robert's definition of ritual as "day-to-day essentials." Scholars have long discussed the fact that newspapers play a societal role, but the habitual aspect of newspaper use is what Stone calls one of the less-than-fully-explored central themes of media research (Stone 1987, p. 108). John J. Pauly (1991) notes that the newspaper habit caught the attention of scholars in the 1930s as they first looked at the societal role of the press. A major train of thought at this time, Pauly said, concentrated

on the role and responsibility of readers in democratic societies. He quoted journalism professor Raymond Nixon as declaring in 1933 The greatest

72 danger to society is not the way papers are edited but in the way papers are read (Pauly 1991, Page 286). Pauly also noted a public effort prior to World War II to instill Americans with the newsreading habit as a patriotic duty. Programs were instituted to tutor readers in higher tastes and in techniques of systematic, critical reading. Educators, journalists, and public officials alike wanted citizens to read more critically, but also more faithfully, in a spirit of civic duty (Pauly 1991, Page 290). An example was was Edgar Dales textbook, How to Read a Newspaper. Pauly said the book was widely adopted by American high schools and had the premise that one must regularly read a newspaper to be a responsible member of democratic society. Often, the ritual-newspaper connection has been examined in terms of the production of news, rather than the consumption of news. Ehrlich wrote that the framework of ritual can be used to show that media workers breathe a journalistic cultural air maintained by common practices in newsrooms. He uses the term ritual to identify the standardized, repetitive actions designed to ease workflow or function such as the reliance on government power structures for news information (Ehrlich 1996). Recently, a few researchers have started to look at the habits of media users. Rothenbuhler makes an eloquent argument for the

73 inseparability of ritual from communication, as a ritual that says nothing is no ritual at all. But his book on the subject, Ritual

Communication, focuses on television and interpersonal communication with little mention of print (Rothenbuhler 1998). Stone and Wetherington (1979) made perhaps the best early effort at quantifying the newspaper habit through a survey of college students in the 1970s. The majority of survey respondents told the researchers they have a particular time and place to read. He also showed a link between habitual reading times or locations and media usage patterns of the parents of the students. Readers of morning and afternoon editions differed on what room they read in, and half the people who said they read a newspaper in the evening identified that paper as the morning Los Angeles Times. Regrettably, the results of the 1979 study cannot be generalized. No attempt was made to obtain a random sample participants were interviewed in the cafeterias and student union of California State University at Long Beach. As college students by definition have a higher educational level than the norm, generally occupy a narrow age cohort and are skewed to the higher end of the economic scale, the study was susceptible to influences of education, age and income.

74 Stone was nevertheless later able to piece together a scenario of newspaper-reading-habit acquisition from the various other readership studies on file. His six stages were: 1. Perusing the Sunday comics prior to first grade. 2. Gaining an interest in content from ages 6 to 12. Sports pages and advice columns are early targets. 3. High-school-age young people set their newspaper reading pattern by making a choice of daily readings (or no reading). 4. Reading increases among young adults, 18-24, especially those who did not pick up the habit earlier. 5. Early adulthood, marked by marriage, starting a household and beginning a family, are associated with higher newspaper use. 6. Patterns remain fixed through the 55-to-65 year old range. Reading frequency drops after age 65, possibly due to health, eyesight or finances. (Stone 1987, pp.108-109) Perhaps the most telling studies on the newspaper habit deal with the time of day at which people read their newspapers. Several researchers studied the late 20th century trend among newspapers to convert from evening (or "p.m.") distribution to morning (or "a.m.") distribution. When he studied the 1982 conversion of the Jonesboro, AR, Sun, Fowler (1985) determined that readers merely changed their habits. Rarick and Lemert's 1983 study of the Eugene, OR, Register-Guard also found a switch in reading times, but found that overall media used changed rather dramatically. Former evening readers spent less total time reading and increased their use of electronic media.

75 Stone (Stone 1986; Stone and Stone 1990) conducted a larger study of merger of the Memphis, TN, Press-Scimitar into the morning Commercial Appeal and took a somewhat more detailed approach. Stone surveyed a large group of readers two weeks before the shutdown of the evening paper, then again three months later and a third time nearly a year later. After a year of receiving a morning paper, a majority of the former evening subscribers still read their newspaper in the afternoon or evening. Several other studies concentrated on whether regular newspaper reading the habit was picked up from ones parents or affected by lifestyle changes (Stone and Windhause 1983). Crompton took this type of research in a somewhat different direction by defining the emotional factors that Canadians attribute to their reading habits. The data she analyzed from the International Adult Literacy Survey showed that people of all educational levels "are equally dedicated to their (reading) habit" but that people with higher levels of education more frequently read for "leisure." These leisure readers are immersed in books, magazines and newspapers for almost an hour and half daily though women are more likely to find comfort in books while men find comfort in newspapers. (Crompton 1997, pp. 11-12) Research into newspaper reading habits are rife with challenges, as Barnhurst and Wartella (1991), demonstrated. They cite a number of studies showing the decline of newspaper readership, noting especially

76 that in 1970 newspaper circulation fell below the number of households in the nation for the first time this century. And their study of undergraduate college students showed that television is now a major factor in the lives of Americans, with respondents spending three times as much time daily watching television than reading the newspaper. (Barnhurst and Wartella 1991, Page 196) Nevertheless, this study of the progeny of the television age showed a lingering appreciation for newspapers and a begrudging acknowledgment of newspaper use by young people. They asked high school students to write life histories of their media experience. Although most initially denied much contact with papers, nearly half the students studied by Barnhurst and Wartella eventually said the newspaper was a constant part of the household background. It was read, but also performed a variety of non-news functions for these families a source of art projects, a focus for family time, an object of entertainment, a patching material for shoes with holes and a cudgel for hitting the dog. As they entered adulthood, many of these young people formed strong opinions about the value of newspaper reading. Their essays often associated quality or education with newspaper reading: It sounds a bit strange, but I think of an elite group when I picture someone reading a newspaper. I feel that more educated people read papers while the less educated just sit back and watch the news on television (Barnhurst and

77 Wartella 1991, Page 203). Like Berelsons study of the New York strike, then, the work of Barnhurst and Wartella shows that newspaper use sometime seems to defy logical explanation. Still, habit was defined as the repetitive reading of the newspaper, not the how and where one reads.

Popular Literature

References to newspaper reading habit are somewhat easier to find (and much more colorful) in the literature of criticism and popular culture. Almost 100 years ago, social psychologist Delos F. Wilcox blasted the newspaper reading habit as a menace to society: ... we must deplore, and so far as possible, overcome the evils of habitual newspaper reading. These evils are, chiefly, three: the waste of much time and mental energy in reading unimportant news and opinions, and premature, untrue, or imperfect accounts of important matters; second, the awakening of prejudices of commercial greed of newspaper managers; third, the loading of the mind of cheap literature and the development of an aversion for books of sustained thought. Thus the daily newspaper often attempts to make the intellectual life of its readers one continuous series of petty excitements, a veritable life of the social "senses," and to shut their minds, by a mere fullness of occupation, against any appeal that does not find a voice in the daily news sheet. (Wilcox 1900) Seymour Krim, writing many years later in The Nation, was less vitriolic but also bemoaned the mental toll he paid to keep up his daily habit of reading the hefty New York Times:

78 Few regular readers have the chutzpah to snub the paper entirely for a day, never doubting they will be punished in some mysterious way, yet few can get through it without psychic confusions about what to skip, whether to read on at the expense of earning a living, vacuuming the rug, writing a letter to the phone company, etc. I kid you not, it's become a weird confrontation for many people, the New York Times, all of its sheer mass comprising nothing less than an alternate reality but fearfully much for the people who don't want to devote their lives to it, as James Joyce asked his readers to do with his books. (Krim 1988) More often, popular references to the newspaper reading habit are couched in terms of self-deprecating humor. Commentators freely admit that they are hooked on reading, but offer only verbal shrugs in explanation. Steve Lopez, writing in Sports Illustrated, took that tack when he described his "agate addiction" his fascination with baseball box scores: I had the newspaper opened to an entire page of baseball box scores when my wife walked into the room, looked at me as if I were a moron and left. I'm not sure how long she was gone maybe five minutes, maybe 25 but when she returned, I had not moved, and she asked how many hours of my life I'd frittered away studying those endless columns of microscopic numbers. Then she said, "The really frightening thing is you look like you're reading Moby Dick. (Lopez 1997) The popular literature also shows a concern for the future of the newspaper reading habit as an icon of American culture. Approximately 700 newspapers in the United States have joined newspapers from 30 nations worldwide that offer Newspapers in Education (NIE) programs. NIE is an educational partnership between the newspaper industry and

79 participating school systems started in the 1930s. It gained new popularity in the 1980s when publishers noted a slip in readership by young people. Today the program is supported by both the World Association of Newspapers and the Newspaper Association of America (WAN 2000). Joe Salzman, a journalism professor at the University of Southern California and a columnist for USA Today Magazine, articulated the concern of both academics and journalists in a 1995 column on the newspaper-reading habit. He noted that for generations of Americans "a habit ingrained in children by their parents was the daily ritual of reading the newspaper" but that the advent of entertaining radio and television has nudged out that habit in favor of the ritual of watching "soaps and MTV, talk and cartoon shows" (Saltzman 1995, p. 39). Without the newspaper reading habit, Saltzman said, the future of the United States itself is threatened: This is a nation ripe for demagoguery because demagogues flourish in an uniformed environment filled with slogans, ignorance, and fear. A nation that has forgotten the habit of wanting to be informed and the habit of taking the time to be informed is a democracy in trouble. It's a country filled with people willing to accept public policy decisions based on misinformation and distortions. It just becomes too hard to take the time to figure out what's right or wrong. It is so much easier to get angry at cardboard characters fighting it out on "Melrose Place." (Saltzman 1995, p. 39) Scholars have shown with little doubt that habit and ritual play important roles in human lives, and even non-scholars

80 have noticed the special connection some readers have with their newspapers. The primary theories of newspaper readership, however, say very little about the impact of habituation on readership or on what stimulates the reading habit. The opportunity for new research in this area of audience study is high, as the following chapters demonstrate.

81

CHAPTER V

PRELIMINARY STUDIES

Two preliminary studies were conducted by the author and contributed substantially to the development and framing of the research in this dissertation. The first study was a master's thesis completed in May 1990 at the University of Texas. The thesis, "Lifestyle and the Daily News: A Comparison of Newspaper Non-Readers, Infrequent Readers and Frequent Readers," represented the author's first substantial effort to quantify the differences between people who ardently read a newspaper and those who read it less frequently or not at all. It attempted to link the desire and passion to read a newspaper to psychographic and demographic characteristics. The second study was completed in April 1998 and represented a very different approach to the same research question. The study, "50 Years Later: What it Means to Miss the Newspaper," was a qualitative investigation of the impact on readers of newspaper delivery failure. Through in-depth interviews, the author sought clues to what raised the anxiety level of people whose newspaper delivery was temporarily delayed. In that review of anxiety, the study also revealed evidence of

82 what factors both content, emotional and personal contributed to the respondent's desire to read a newspaper. This chapter will summarize the research and findings of those two studies to provide background for the primary research in this dissertation.

1990 Study:

Introduction

The 1990 study was undertaken in an attempt to understand what differentiates people who frequently read a daily newspaper, those who only infrequently read a daily newspaper and those who do not read a daily newspaper. It came during an academic break from the author's lengthy career in newspapers and was a direct result of the professional curiosity about what drives readership mentioned in Chapter 1. The study examined both demographic and lifestyle characteristics and their relationship to newspaper readership. It found that there are

several significant demographic indictors of newspaper readership but that lifestyle is a less certain indicator. Like many other journalists, the author was trying to make sense of declining newspaper readership in an era that placed increasing value on information. Countless theories had been offered to account for this phenomenon. Rykken (1989) said newspaper publishers have blamed the slippage on

83 time pressure on readers, distribution difficulties, increased competition from other media, the changing roles of women, illiteracy, a busier social calendar, a mobile society with few ties to a community, and little interest in local news. Westley and Severin (1964) linked non-readership to age, income, occupation and social isolation. Both Manheim (1976) and Sohn (1975) lay the blame on the proliferation of television. This study, however, explored the possibility that newspaper readers hold a value set different from that of infrequent readers. The study focused on newspaper readers, but with the realization that any investigation of the habits of that population must be made in the context of the use of other media. Other media use, particularly television viewing, has been cited as a primary cause for the decline in newspaper readership in Americans. Sohn (1975) lamented that the proliferation of television is turning the United States into a nation of "vidiots." It might also be argued that the use of other media is an effect a response to content in newspapers that fails to appeal to the values and interests of potential readers. With that in mind, newspaper consumers and television audiences were compared, as were non-readers of newspapers with non-viewers of television news. Of special concern was the consumption habits of "infrequent" readers, as the literature historically showed they represented a better

84 opportunity for newspapers to increase their circulation. The National Advertising Bureau estimated 28% of American adults were infrequent readers of newspapers, while only 10% were non-readers (NAB 1985). However, a similar study reported by the American Association of Newspaper Editors in 1998 estimated 24% of American adults were "occasional" readers and 25% were non-readers (ASNE 1998). A stumbling block on the comparison of reader frequency studies has been the inconsistency in defining the terms. Penrose, Weaver, Cole and Shaw, for instance, labeled as a "nonreader" anyone who answered negatively to "Have you read a newspaper yesterday or today?" (Penrose et al. 1974) Wanta, Wu and Hu called anyone who read less than two days a week a nonreader (Wanta, Hu, and Wu 1995). The NAA and the ANSE

(who shared a data source) said anyone who read one to three days in the past five weekdays was an "occasional" reader (NAA 1998, p. 83). Nevertheless, infrequent readers are already familiar with the product the newspaper and just have to be persuaded to use it more often. If it can be shown that infrequent readers have a need for the information provided by a newspaper (for instance, if they are high consumers of news from other media) it might be argued that they simply have a poor attitude about newspapers. Marketing research shows that consumer attitude is much easier to change than consumer need (Assael 1987).

85 The differences between frequent and infrequent readers is not easily linked to demographics. Although women read less frequently than men, the trend appears to be related to a shortage of discretionary time rather than attitude. Homeowners read more frequently than renters, but the only significant correlation between civic involvement and reading frequency is voting (ASNE 1998, p. 2.4) This lack of a clear demographic "magic bullet" has led several researchers into the less clearly defined field of psychographics, a market research tool pioneered in the 1960s by Emmanuel Demby (Weinstein 1987) Weinstein argued that psychographics technically relate to consumers' personality traits, while lifestyle consists primarily of individuals' attitudes, interests and opinions ... known among market researchers as "AIOs." However, he noted that in practice lifestyle is the more useful factor and is considered collectively with personality traits, becoming a part of practical psychographics. The term "lifestyle" itself grew out of the 19th century sociological studies of Max Weber and later of psychiatrist Alfred Adler, both of whom saw an individual's style of life as a motivating force (Schweitzer 1977). Newspapers editors, publishers, circulation managers and advertising managers, said Schweitzer (1977, p. 1), increasingly turn to market research for answers to the perplexing riddle of flat circulation growth at a time of increasing population. Lifestyle analysis is one method

86 that might suggest solutions to circulation and readership problems. If newspapers can be designed to appeal to what people want to read, the notion goes, people will buy them and read them. This is the marketing concept. How valuable are demographic and lifestyle comparisons of frequent and infrequent newspaper readers? Considering the concern among newspaper publishers about the declining percentage of American's reading newspapers, they could be quite valuable. Tipton's (1978) review of readership studies for the American Newspaper Publishers Association said there is little hope of attracting chronic non-readers but that occasional newspaper readers occupy a market segment that can be brought into the fold of regular readership. Much later, the ASNE readership report label 15% of American adults as "prime prospects" because they not only fit the demographic profile of "hard core" readers, but they occasionally read the paper now (ASNE 1998). Tipton further noted that the demographics of readership are relatively consistent: Readers are older, more highly educated and have higher incomes. Non-demographic factors such found in lifestyle research, particularly interest in politics and community are more related to newspaper readership than such demographic factors as age or education, he said.

87 Clearly, then, there are generalizable differences between regular readers of newspaper and infrequent readers of newspapers.

1990 Study: Hypotheses and Methodology

The 1990 study of potential readers in Austin, Texas, focused on nine hypotheses: 1. Infrequent readers of newspapers will fall into different lifestyle groups than frequent readers or non-readers. 2. Infrequent readers of newspapers will fall into different demographic groups than frequent readers or non-readers. 3. The more people read a newspaper, the more liberal they will tend to rate themselves. 4. The more people read a newspaper, the less they will value family life. 5. The more people read a newspaper, the more they will watch television news. 6. The greater people's income, the more likely they are to frequently read a newspaper. 7. The older people are, the more likely they are to frequently read a newspaper. 8. The more highly educated people are, the more likely they are to frequently read a newspaper. 9. Women are more likely to be infrequent readers of newspapers than men are, but less likely to be non-readers of newspaper than men are. The study was part of an omnibus telephone survey of culture and communications undertaken by graduate students both at the University of Texas at Austin, Texas, and at San Jose State University in San Jose, Calif., during the spring semester of 1989. The UT study was coordinated by Dr. Pamela Shoemaker. The California study was coordinated by Dr. Diana Stover Tillinghast.

88 Telephone interviews for the Texas survey were completed in March and April 1989 on a probability sample of 349 telephone numbers in the Austin, Texas, local telephone dialing area, which had an estimated population exceeding 600,000. The telephone interviews for the California study were completed in the spring semester of 1989 on a probability sample drawn by random digit dialing from the San Jose local telephone dialing area. The San Jose primary metropolitan statistical area had a population of 1.07 million. To avoid bias from the large number of unlisted telephone numbers in the area (53,200, or approximately 21 percent of the 253,700 working residential telephone numbers in the Texas study), phone numbers were generated by the Waksberg technique of random digit dialing (Waksberg 1978). Eligible respondents were residents of the Austin dialing area, at least 18 years old and who said they live in households other than group homes, prisons, fraternity houses and other non-traditional households. The UT survey produced 349 valid responses, for a response rate of 71.4 percent by the procedural formula, or 49.6 percent by the more standard formula of the Council of American Survey Research Organizations (CASRO). The Texas questionnaire included 99 questions, of which 58 were analyzed for this study. The results of this survey were combined with

89 those of a California survey that contained many of the same questions. The combined results produced 908 valid cases. The 48 non-demographic questions were considered lifestyle questions, under the explanation of Levy's delimiters of lifestyle by Robertson (Robertson 1970) value questions, interpersonal attitude questions and daily routine questions. For analysis, 20 questions were grouped under the "media use" label, and 15 were "social values," and 13 were "family values." The media use questions started by asking "Where do you get most of your news and public affairs information from newspapers, radio, TV, magazines by talking to other people or where?" Respondents where then asked if their interest was primarily in local or national news. Respondents were asked how many days a week they read a daily newspaper; how many minutes each day; and how frequently they read local, national and world news and editorials. A similar set of questions probed the usage of television by the respondents. A third set of questions queried respondents about radio use. The social value questions were similar to the family value questions, but on a variety of topics. Most were Likert-type questions employing the five-point disagree-agree scale. Examples are "In American society, connections or who you know is very important," "A husband and wife who don't get along and have children should get a divorce" and "In

90 choosing the person you will marry, love is more important than anything else." The family value questions were scattered throughout the questionnaire. The primary measure of family value was a question asking "How important is it for you to have a warm and close family?" Respondents were asked to answer on a four-point scale of (1) not at all important, (2) important, (3) somewhat important and (4) very important. Respondents were also asked closed-ended questions about whether they get together with their family other than at meals and if so, how often and what do they do while they are gathered. Traditional family values were tested by a series of Likert-type questions to which respondents were asked whether they (1) strongly disagreed, (2) disagreed, (3) neither agreed nor disagreed, (4) agreed or (5) strongly agreed. Samples of these questions are "Aging parents should be taken care of by their children" and "Younger people should show proper respect to older people." Data from the survey were analyzed with the SPSS-X statistical program. First, frequencies were determined for all questions. From the frequencies of the media use questions, boundaries for "regular" and "infrequent" media users were determined by dividing the groups at the mean.

91 Cross tabulations between frequent and infrequent users were run on all categorical questions. The ANOVA function was used to analyze questions that produced other data.

1990 Study: Key Definitions

Readership for this study was measured by asking respondents "About how many days a week do you read a daily newspaper?" Answers were given on a scale of 0 to 7. Respondents who answered 0 were classified as non-readers; those who said answered 1 to 6 were classified as infrequent readers and those who answered 7 were classified as frequent readers. Television news viewership was measured in two stages. First, respondents were asked "About how many days per week they watched local news." Answers were given on a scale of 0 to 7. Then they were asked "About how many days per week do you watch national news?" Again, answers were given on a scale of 0 to 7. The two scores for each respondent were combined. The resulting score on a scale of 0 to 14 was used as the measurement for television news viewership. Respondents whose score was 0 were classified as non-viewers. Those whose combined score was at or below the mean of 7 were categorized as infrequent viewers. Those who were categorized above the mean of 7 were categorized as frequent viewers.

92 Overall news exposure was measured by combining the scores for newspaper readership and television news viewing, creating a nine-point typology. Media use was measured by 16 questions, with the intent that most of these would be combined into an additive index of media use. For the purposes of creating the index, all variables were standardized. Cronbach's Alpha was calculated to asses the reliability of the index. Lifestyle was measured by 15 questions on social values. Of those questions, 14 had responses on a five-point scale where 1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, and 5 = strongly agree.

1990 Study: Results

The combination of the Texas and California studies produced 906 valid cases for analysis, although listwise deletion of missing data due to unanswered questions by some respondents decreased that number for some tests. Table 2 shows the distribution of cases by two measures of newspaper readership: Number of days per week respondents said they normally read a daily newspaper and the collapsed scale of non-readers, infrequent readers and frequent readers. Non-readers were those respondents who, when asked how many days a week they read a

93 newspaper, answered "0." Those who answered "1" to "6" were classified as infrequent readers and those who answered "7" were classified as frequent readers. The respondents clustered heavily on both ends of the 0-7 readership scale, with 33.7% saying they never read a daily newspaper and 42.6% saying they read every day. Of those in between, the highest percentage was for five days a week, which accounted for 8.4 percent of the sample. TABLE 2. Newspaper Readership by Frequency Question How many days a week do you read a newspaper? Value Frequency 0 1 2 3 4 5 6 7 305 10 14 47 31 76 37 386 906 Percent 33.7 1.1 1.5 5.2 3.4 8.4 4.1 42.6 100.0

n=906 Mean = 1.3 Standard Deviation = .47 Total Newspaper readership (N=906) Non-readers (0 days) Infrequent readers (1-6 days) Frequent readers (7 days) Total

305 215 386 906

33.7 23.7 42.6 100.0

94 Using the collapsed scale, 33.7% were non-readers, 23.7% were infrequent readers (at least one but less than 7 days a week) and 42.6% were frequent readers (7 days a week). A mere 7.1% of the sample said they never watch television news. However, only 10.2 percent said they always watch both national and local news. On the collapsed scale, 7.1% were non-viewers of television news, 48.8% were infrequent viewers and 44.1% were frequent viewers. Combining the television news viewership and the newspaper readership statistics produced a nine-cell typology illustrated in Figure 3. Only 2.8% of the respondents were truly averse to news they neither read newspapers nor watched television news. quarter of the respondents (24.4%) Conversely, nearly a

were "news junkies" who rated

themselves both frequent viewers of television news and frequent readers of newspapers. Data collected from psychographic and demographic questions generally approximate the national statistics of the time cited by Bogart (1988): A population that is middle aged, well educated, fairly affluent and has lived in the same area for a considerable time.

95

Newspaper reading
None 1-6 days 7 days

1
Non-viewers/ non-readers

2
Non-viewers/ infrequent readers

3
Non-viewers/ frequent readers

N=61

n=24 Television viewing

n=10

n=26

4
Infrequent-viewers/ non-readers

5
Infrequent-viewer/ infrequent readers

6
Infrequent-viewer/ infrequent readers

N=421

n=168

n=119

n=134

7
Frequent-viewers/ non-readers

8
Frequent-viewers/ infrequent-readers

9
Frequent-viewers/ frequent readers

N=380

n=97 n=305

n=73 n=215

n=210 n=386 n=908

FIGURE 3. Comparison of newspaper readers and television viewers

96 The data for media use showed that newspapers barely edged out television as the primary source of news, with 41% of the respondents indicating they get their news from newspapers and 38.2% saying television was where they get their news. Radio was cited by just 13.6%. Most of the subjects (66.4%) reported they read a daily newspaper regularly. This is consistent with contemporary statistics compiled by the Newspaper Advertising Bureau (1986, p. 16), which estimated that 64% of American adults read a newspaper on the average weekday. The

earlier NAB study also showed Americans typically spent 44 minutes a day reading a newspaper (NAB 1985). Respondents to this study spent and average of 36 minutes reading a newspaper. The same NAB study showed 67% of American adults watch some television news on a typical day. In this study, 85.2% of the respondents said they watch local television news and 77.6% said they watch national television news. There was a high level of agreement among respondents to a social value statement "You should speak your own mind, no matter what other people think." But when given the statement "When I disagree with my boss, I tell him or her," the level of agreement was quite low. It seems that free thinking is a popular concept when stated as a theory, but much less popular when tested in certain lifestyle situations.

97 Politically, the survey participants rated themselves slightly on the conservative side of the continuum. Asked about their work ethic, nearly 62% said they work more than others do, and only 18.5% said they work as much as they feel like working. These data were reanalyzed to compare the appropriate mean scores or percentages of newspaper non-readers, infrequent readers and frequent readers on the 58 questions. Interval data questions were analyzed by ANOVA, and non-continuous data questions were analyzed by crosstabulation and Pearson r correlation. This comparison produced correlations significant at the p<.05 level for 21 variables, pointing to interesting differences between the sample groups. In demographics, frequent readers both had lived in the survey area longer (p<.001) and were significantly older than either non-readers or infrequent readers (p<.001). The mean age of frequent readers, 43.85 years, was nearly a decade older than either of the other two groups. This supported Hypothesis 7. Education showed a strong positive correlation with newspaper readership (p<.001). The more educated respondents were, the more likely they were to frequently read a newspaper. This supported Hypothesis 8. Income also showed a positive correlation with newspaper readership (p<.001), with infrequent readers being more affluent than

98 non-readers, and frequent readers considerably more affluent than either. This supported Hypothesis 6. Among respondents who lived with others, frequent readers were significantly (p<.05) more likely to live with people they considered "family" than those they just considered "roommates." This seems counter to Hypothesis 4, which stated that the more people read a newspaper, the less they will value family life. Gender was also significantly correlated with readership (p<.05). Although infrequent and frequent readers were nearly evenly divided between males and females, females were much more likely to be nonreaders than were males (Table 3). Among non-readers, 59.3% were women and 40.7% were men. This is counter to Hypothesis 9. Marital status was a statistically significant indicator of readership (p<.001) that gave distinction to the infrequent readers. While nonreaders were somewhat more likely to be married than single (46.5% married, 36.5% single) and frequent readers were much more likely to be married (64.1% married, 21.4% single), infrequent readers were more likely to be single (39.3% married, 48.5% single). Frequency and percentages for demographics questions by nonreaders, infrequent readers and frequent readers were much more likely to be married (64.1% married, 21.4% single), infrequent readers were more likely to be single (39.3% married, 48.5% single).

99 TABLE 3. Frequency and Percentages for Demographics Questions by Non-readers, Infrequent Readers and Frequent Readers
Non-readers Question Value Do you live alone or with other adults or children? X2 = 2.28 df = 2 Cramer's V = .05, p >.05 Do you consider the people you live with to be family or are they just roommates? X2 = 6.90 df = 2 Cramer's V = .10, p <.05 Are you male or female? X2 = 8.19 df = 2 Cramer's V = .95, p <.05 What is your marital status? X2 = 60.32 df = 8 Cramer's V = .18, p <.001 Freq % Infrequent readers Freq % Frequent readers Freq %

Alone With others Totals

73 231 304

24 76 100

45 168 213

21.1 78.9 100

74 310 384

19.3 80.7 100

Family Roommates Totals

198 32 230

86.1 13.9 100

144 24 168

85.7 14.3 100

285 24 309

92.2 7.8 100

Male Female Totals

124 181 305

40.7 59.3 100

107 107 214

50 50 100

197 189 386

51 49 100

Married Single Divorced Widowed Other Totals Totals

140 110 38 9 4 301

46.5 36.5 12.6 3 1.3 99.9

84 104 21 5 0 214

39.3 48.5 9.8 2.3 0 99.9

243 81 32 19 4 379

64.1 21.4 8.4 5 1.1 100

What is your occupation? X2 = 68.68 df = 18 Cramer's V = .20, p <.001

Professional Managerial/Exe cutive Sales and Service Clerical Skilled Labor Unskilled Labor Housewife/Hus band Student Retired Unemployed Totals

92 22 27 30 32 12 25 37 14 2 293

31.4 7.5 9.2 10.2 10.9 4.1 8.5 12.6 4.8 0.7 99.9

55 14 29 24 24 14 9 33 4 6 212

25.9 6.6 13.7 11.3 11.3 6.6 4.2 15.6 1.9 2.8 99.9

115 24 36 25 50 16 36 16 49 5 372

30.9 6.5 9.7 6.7 13.4 4.3 9.7 4.3 13.2 1.3 100

100 A comparison of the use of various parts of the newspaper by infrequent readers and frequent readers produced intuitively predictable results: Frequent readers read all parts of the paper more often than do infrequent readers. Even though one might intuitively think non-readers of newspapers would be the greatest fans of television, it was not always so. Frequent readers of newspapers said they watched an average of 2.53 hours of television daily, compared to only 2.49 hours for non-readers and 2.31 hours for infrequent readers. Although this relationship was not statistically significant, viewing did become significant (p<.05) when narrowed to how often respondents watched national news on television. Frequent readers watched far more national television news than either non-readers or infrequent readers, reporting a mean of 3.74 viewing days a week compared to 2.72 for nonreaders and 2.60 for infrequent readers. This partially supports Hypothesis 5, which stated that the more people read a newspaper, the more they will watch television news. The anomaly is the infrequent reader, who also watches the least television in general and national news in particular. It may be an indication that lack of time is an essential factor for the infrequent reader, especially when one notes that infrequent readers listen to significantly more radio (p<.05) than do non-readers or infrequent readers. Radio is

101 one medium that can be consumed while doing something else, particularly while driving a car. When one looks at where the respondents get their news, the intuitive theory holds true. Non-readers, to no surprise, read less than infrequent and frequent readers and more frequently count on television and radio for their news. (Curiously, however, 6.2% of those who said they do not read a daily newspaper also said the newspaper was their primary source of news.) The pattern that seems to emerge from this data is those who enjoy news seek it in any medium. The social value questions produced the fewest correlations with readership. Only one question: "In choosing someone to marry, love is more important than anything else," produced a significant correlation (p<.05). On that questions, non-readers and frequent readers had nearly identical mean scores (3.93 and 3.92) indicating they agree with the statement. Infrequent readers had a mean score of 3.76, moving them closer to "neither disagree nor agree." This single correlation was not enough, however, to support Hypothesis 1, which stated "Infrequent readers of newspapers will fall into different lifestyle groups than frequent readers or non-readers." While there was variance among the three groups on the political ideology measure, it was not significant, failing to support Hypothesis 3.

102 As with the social value questions, the family value questions failed to produce significant correlations with newspaper readership. Mean scores on the question "How important is it for you to have a warm and close family?" were nearly identical among the three groups of respondents, with family value being important to everyone. This fails to support Hypothesis 4.

1990 Study: Review of Hypotheses

Hypothesis 1: Infrequent readers of newspapers will fall into different lifestyle groups than frequent readers or nonreaders. Using the operational definition of "lifestyle" employing the social value questions, this hypothesis was not supported. Only one of the questions produced a significant correlation, and this could be attributed to Type 1 error. Hypothesis 2: Infrequent readers of newspapers will fall into different demographic groups than frequent readers or nonreaders. This hypothesis was supported. Infrequent readers were younger and much more likely to be single than frequent readers or non-readers. Hypothesis 3: The more people read a newspaper, the more liberal they will tend to rate themselves. This hypothesis was not supported, as the political ideology question failed to show a significant correlation with newspaper readership.

103 Hypothesis 4: The more people read a newspaper, the less they will value family life. This hypothesis was not supported, with some evidence pointing to the opposite effect. Although a direct question on the value of a "warm and close family" did not produce significant results, frequent readers were much more likely to be married than single, divorced or widowed. In addition, they were much more likely to consider the people with whom they live "family" than just "roommates." Hypothesis 5: The more people read a newspaper, the more they will watch television news. This hypothesis was supported. Nearly a quarter of the respondents were both frequent newspaper readers and frequent television news viewers. Also, newspaper reading and television news viewing were significantly correlated to each other. Hypothesis 6: The greater people's income, the more likely they are to frequently read a newspaper. This hypothesis was supported. Frequent readers represented the only group with a mean income above $40,000. Hypothesis 7: The older people are, the more likely they are to frequently read a newspaper. This hypothesis was supported. Hypothesis 8: The more highly educated people are, the more likely they are to frequently read a newspaper. This hypothesis was supported.

104 Hypothesis 9: Women are more likely to be infrequent readers of newspapers than men are, but less likely to be nonreaders of newspaper than men are. This hypothesis was not supported by the data. Gender showed a negative correlation with newspaper readership. Women were far more likely to be non-readers than men were but just as likely to be infrequent readers or frequent readers.

1990 Study: Discussion

This research project illustrates well the frustration of newspaper audience studies. It was clear to the author intuitively and then shown quantitatively that newspapers hold a different appeal to different types readers. Determining what causes that difference is far less clear, however. This study, like others before it, pointed to demographics as the most reliable indicators of newspaper readership. Older, more affluent and more educated people are much more likely to regularly read a newspaper than are the young, the poor and the less educated. Knowledge of these facts is of only limited help, however. Does a publisher simply "write off" those who do not fit the newspaper's "best fit" demographics? If the industry tries to attract new readers, what editorial themes and marketing strategies are likely to appeal to what newspaper

105 consultant Conrad Fink (Fink 1989) called the "unwashed and untutored"? The answer to the first question is most certainly "no." The 1990 study found frequent readers had a mean age of over 40, a family income over $40,000 per year and at least some college education. However, according to the U.S. Census Bureau (1987) statistics for the time, only 37.1% of Americans had a family income of more than $35,000; only 31% of American adults were at least 40 years old; and only 35.7% of Americans had at least some college education. To follow the "best fit" strategy, then, would mean writing off nearly two-thirds of American adults. Current concerns for declining readership pale in comparison and elimination of that many would threaten the newspaper's status as a mass medium. But of greater concern is the fact that the statistics simply don't make sense. While only a third of American adults may fit the "reader profile" of demographic research, more than half actually do read the paper on a regular basis. The author's 1990 discussion of the survey results found great hope in the finding that the various news media complement one another more than they compete with one another. Frequent newspaper readers are also frequent viewers of television news and frequent listeners of radio news. News, then, is the valued commodity, not newspapers or television

106 broadcasts. The author's advice at the time was that publishers and editors ask themselves what newspapers can provide that will turn casual interest into a craving for news. The challenge, the study concluded, is to honestly repeat the first sentence of the 17th century description of newspapers by English essayist Charles Lamb without giving a new, negative, meaning to the second: Newspapers always excite curiosity. No one ever lays one down without a feeling of disappointment (Lamb 1833).

1998 Study:

Introduction

The second preliminary study, completed in 1998, explored modern corollaries to Lamb's observation through qualitative methods. "50 Years Later: 'What It Means to Miss the Newspaper'" revisited a landmark investigation of newspaper readership published a half century ago in an attempt to discover why regular newspaper readers miss their paper when they cannot receive it. This study used individual stoppages caused by normal delivery problems at a daily newspaper in lieu of the strike that prompted the original study In the 1948-49 edition of Communication Research, Bernard Berelson of the Bureau of Applied Social Research at Columbia University published a study that was cited in textbooks and other research papers for the next five decades. What Missing the Newspaper Means was an

107 analysis of the 1945 New York newspaper delivery strike through the eyes of the would-be readers who didnt get their usual daily paper. Other researchers also focused on the strike, but in a more quantitative manner that tallied the ways people were keeping up with the news. Berelson, however, had an educated hunch that missing the newspaper was less a statement of physical loss than it was of social and psychological trauma. He wanted to know what people felt when the paper didnt arrive, and why.

1998 Study: Berelson's Research

When the delivery workers of eight major New York City newspapers went on strike the afternoon of June 30, 1945, Berelson was curious. A week later, when New Yorkers still had no printed news and two scientific polls by national organizations appeared to be unable to address the attitudes of denied readers, he and his researchers went into action. The strike lasted 17 days, and was so effective at stopping circulation that most New Yorkers were effectively deprived of their regular paper. By going to a newsstand, they could buy one exempted newspaper, PM, and a few specialty publications. They could also buy copies of their regular paper at the counter of the papers central office, but the vast majority of Gotham readers were left paperless (Berelson 1949, p.

108 111). The Columbia researchers under Berelson elected to use what he called an exploratory survey that was quite different from the more traditional surveys launched by the Elmo Roper agency and by Fact Finders Associates, Inc. Berelson said that his goal was to understand the function of the modern newspaper for its readers: Where the Roper and Fact Finders surveys were extensive, the Bureaus was intensive, designed to secure psychological insight in order to determine just what not having the newspaper meant to people (Berelson 1949, p. 112). The strike offered an ideal laboratory, he said, because people are more conscious of what the newspaper means to them during such a shock period than under normal conditions. In addition, a crisis tends to make them more articulate about such matters (Berelson 1949, p. 112113). Berelson used a questionnaire to frame interviews with 60 people scattered through the rental areas of Manhattan. While noting that it did not employ a statistically reliable sample, Berelson said he felt the technique allowed him to get beneath the surface facts that the other two polls targeted. There were, of course, a number of responses that paralleled the traditional surveys performed by the Elmo Roper agency and by Fact Finders Associates, Inc., not to mention earlier research and surveys taken in the 50 years since. Like the two other studies, Berelson found readers

109 turned to the radio (there was no commercial television at the time) for news in lieu of their normal newspaper. But chief among the welldocumented truisms of media research confirmed in all three studies was that practically everyone pays tribute to the value of the newspaper as a source of serious information about and interpretation of the world of public affairs (Berelson 1949, p. 114). He said that virtually all of his respondents lauded the paper for its informational and educational aspects. However, Berelson found that his research echoed earlier studies in showing that not nearly as many people depended upon the paper for specific news as the number who claimed this was important. Very few respondents said they wanted a paper to keep up with the hot stories of the day, while many more responded with a variation on the cliche to keep informed. From the comments he collected, Berelson, was able to identify five special uses for newspapers that regular readers seemed to desire: 1. Information and Interpretation of Public Affairs: The serious side of newspaper coverage. 2. Tools for Daily Living: The many lists, rosters, regular features, advertisements and calendars routinely published by. 3. Respite: The relaxation or entertainment value of the daily newspaper . 4. Social Prestige: Many readers felt the newspaper was important not just because it gave them information, but because it enabled them to appear more informed at social gatherings

110 5. Social Contact: Human interest stories, personal advice columns, gossip columns and their ilk provided much more than respite from daily routine. Many people, he found, used them as guides to prevailing morality, insight to the private lives of others and indirect personal contact with distinguished people. (Berelson 1949, pp. 114-120) Even after he had pulled together this list of reported newspaper attributes, Berelson found something missing in his analysis. Where the direct answers to his questions showed categories of desire, the follow-up conversations showed a less specific attraction to the newspaper. He attempted to find answers in psychology. The psychology literature of the time gave him some insight, but it was so marginal he relegated it to just a footnote: When talking about how they missed the

paper, many respondents mentioned a glass of water, a cup of coffee, an appetizer, a piece of candy or something else they would put in their mouth. Berelson found a tie to these statements with Margaret A. Ribbles

writings on the oral fixation of infants. From his writing, Berelson obviously was not convinced that this was firm evidence of reading as a socially acceptable source of oral pleasure. He conceded, however, that newspaper reading may serve the function of a pacifier for adults (Berelson 1949, p. 114, p. 123). As he found no published theory to rely upon, Berelson turned to the comments of his respondents and catalogued them under his own list of non rational uses. These, he said, should be the basis of further more

111 intensive research. The context of the newspaper creates an unusual set of human conditions, Berelson said. It is inexpensive and easy to acquire. Unlike books and magazines, it can be conveniently taken in capsules (read a bit at a time). Of all sources of reading material, it is the most readily available and easiest to consume. The prime emotion these factors provide readers, Berelson speculated, is a sense of security or at least insecurity when deprived of it. One of his respondents said it bluntly: I am like a fish out of water I am lost and nervous. Im ashamed to admit it (Berelson 1949, p. 125). This is closely tied to the ritualistic and near-compulsive character of newspaper reading, though whether the security leads to the ritual or the ritual leads to the security is a chicken-and-egg question. Either way,

Berelson found strong evidence of compulsive or ritualistic behavior among his respondents. Many read their paper in the same way at the

same time every day. At least half the respondents freely referred to the habit of newspaper reading, some using strong descriptions: Something is missing in my life. I am suffering! Seriously. I could not sleep, I missed it so (Berelson, 1949, Page 126). To be fair, not everyone missed the newspaper. A minority of respondents told Berelson they were a bit relieved to be without news of crime, war and mayhem. Nevertheless, Berelson concluded that his study showed that newspaper reading has value per se in our society.

112

1998 Study: Research Questions and Methodology

While the 1998 research project asked the face-value question of Why do people miss their paper?" it was actually a framework for exploring the attitudes of readers toward newspapers. The primary research question, was, Can newspaper readers articulate why they feel a need to read? Because Berelson used a unique situation involving a massive labor strike, this project also has a secondary research question: Does a shortterm stoppage of newspaper delivery stimulate the same crisis-enhance articulation about newspapers that Berelson found? As was true with the original study, this research was based on a limited number of in-depth but timely interviews of regular newspaper readers in a specific geographic area. A daily newspaper in rural eastern Oregon, with an evening circulation of approximately 12,500, agreed to make its daily miss list available to the researcher. This is the list of subscribers who have called the newspaper to report that they did not receive the newspaper as usual that day. Oversights by carriers, counting mistakes in the mail room or accounting errors create misses in the circulation systems of all home-delivered newspapers.

113 Circulation officials at the paper sent an electronic mail message to the researcher early each evening containing the names and telephone numbers of subscribers who called the newspaper office to complain that their newspapers were missing. To make best use of the shock of missing

the paper and to approximate the crisis situation that Berelson said increased the articulation and consciousness of his respondents, the researcher attempted to telephone the subscribers within a short time of their non-delivery complaint to the newspaper. When a subscriber could not be reached that night, she or he was telephoned the next day. Once reached by telephone, the research project was explained and permission was to proceed was obtained from respondents before they were interviewed. Over a period of two weeks March 1998, 35 newspaper subscribers were interviewed by telephone and their comments transcribed. The respondents ranged in age from the early 30s to the late 80s, and included both residents of the larger home city of the newspaper and many of the smaller communities around it. This particular newspaper was selected because it had a high penetration in several of the towns that it serves. The study conformed to the participant suitability standards for qualitative research outlined by Thomas Lindlof in Qualitative Communication Research Methods (Lindlof 1995). Lindlof said interviewees should be selected for their appropriate experience in the cultural scene,

114 ability to articulate their experience in an interview, and ability to devote time to the interview. The fact that the respondents had called the newspaper to report a delivery problem assured a level of appropriateness, the identification and introduction protocols of the Human Subjects Compliance procedure addressed the second factor, and the use of relatively brief telephone interviews conducted early in the evening addressed the third factor. As Berelson did in his study, this project employed a short list of questions to guide interviews with the deprived readers. But also like Berelson, the study counted heavily on the ad lib comments of the participants to illustrate the emotions they felt. Subjects were asked to recreate their emotional condition at the moment they discovered their newspaper was missing, and to speculate what life would be like if a situation came up (like a move to the back country) where no paper would be available. They were also asked about what they read in the paper and why they felt it was important for others to read. The comments were analyzed critically and compared to the findings of Berelson and others. The search for what people missed about their newspapers took on many of the aspects of Erving Goffmans concept of frame, and certainly was well within the bounds of the factist perspective described by Alasuutari (1995).

115

1998 Study: Findings

Berelson might have been very comfortable with the Oregon study, as the small-town readers who responded to it echoed many of the comments of his New Yorkers. Practically everyone contacted paid tribute to the serious news value of the newspaper with a catch. Time and

time again respondents in this study talked about the newspaper as a source of local news. A sampling of the comments: We depend upon it to keep us informed about local news. Not everyone needs it, but we do." Everyone watches TV for national news; its the local news that is important in the newspaper.

This subtle change from a generalized view of news to specifically local news may be attributable to two factors one geographical and one historical. Berelsons subjects lived in one of the largest, most cosmopolitan cities in the United States. Given the size and international impact of New York, it may be difficult to distinguish what local means for those who live there. For rural Oregonians living in towns of a few thousand souls, local news is easily defined as that which involves their friends and neighbors. A second factor is the major change in the news environment since Berelson conducted his study the advent of television. While the only

116 electronic alternative available to Berelsons subjects in 1945 was radio, surveys show that approximately 72 percent of Americans say they rely on television for most of their news (Feder 1997). But the broad geographic area over which a television signal ranges may preclude the detailed local coverage of smaller towns that newspapers are often able to provide. The personal role a newspaper plays in their lives was often best explained by the respondents when they were asked to describe the emotions they felt at the moment they realized their newspaper was not going to arrive this particular evening. The readers expressed a mixture of disappointment and loss, but not anger. A sampling:

I was kind of upset, and wondered what the heck had happened. We look forward to the paper coming. Its kind of like a friend or neighbor coming in. I was frustrated, because I look forward to getting the paper. I mean, part of my nightly ritual is to read the newspaper.

As Berelson found, the crisis (even if a mini crisis) of being paperless seemed to make subscribers eloquent about the role a newspaper plays in their lives. Asked how they would cope if they had to move to a mountain community that had no newspaper delivery, some of the readers were plaintive: We would have to have some sort of paper, even if it came by mail.

117 We could watch TV, but its not like having a paper in the hands. Wed probably have to rely more on radio, but it probably wouldnt be enough for us. "When we got to town, we would just have to buy some newspapers.

A 42-year-old father of two and an avid outdoorsman waxed poetic about his love affair with the newspaper, after explaining that he would go to great lengths to find an edition to read: Let me explain it this way. When I go hunting, I always take four or five newspapers with me up to the mountains. Reading is relaxing, like sitting on the bank fishing. It gives me a chance to unwind. While Berelsons five newspaper attributes public affairs information, tools for daily living, respite, social prestige and social contact seemed to hold in the Oregon study, the notions of ritual and of interactive shared use of the newspaper came to the forefront. For almost all the respondents, reading the evening newspaper was a daily ritual that seldom varied. This was especially true of retired couples, who almost always said they read the paper in the same manner: Each took a section of the paper to their favorite easy chair, read it leisurely, then exchanged sections with their partner. Conversation and commentary on stories was reported as a frequent part of this newsreading ritual. Even

the exchange of sections was a source of entertainment for one couple, as described by a 60-year-old woman who, with her husband, had subscribed for 30 years:

118 We have a great lot of fun sailing it across the room. We laugh about it and tell our friends that we spent the evening throwing the newspaper at each other. About half of the couples said they have a routine in which one partner always grabs a particular section (husbands often went for the sports section), but for others it was first-come, first-served. We just split it up, one man said. We are democratic about it, even though we are Republicans. It depends on who gets it first. A mid-30s mother of two described the daily delivery of the newspaper as an all-family event. My sixth-grade son grabs the front section and reads all the news. My fourth-grade daughter flips through all the sections, looking for names of people she knows. My husband reads the whole paper, cover to cover. I just read the headlines and the obituaries. Im a nurse, so I always look to see if any of my patients have died. Respondents often referred to this social ritual of reading the paper if asked why they couldnt just turn on the television news and do without the paper. Here they were less articulate, but insisted that the newspaper was more to them than information on a page. I just kind of like the whole thing, explained a woman who reads and discusses the paper with her husband every evening. A 73-year-old woman said the paper actually provides a sense of physical comfort for her that she doesnt get from television. The paper doesnt glare back at you, she said.

119 The permanence and physical presence of the newspaper was also mentioned as an important quality. Several people talked of clipping stories and sending them to friends or posting them on their refrigerators. Others said the paper gave them more control over their news intake. 57-year-old man explained this attitude: If the phone rings while I am watching TV news, its gone. But with the newspaper, I can go at my own pace and take all the time I want to digest it and really get meaning out of it. Some readers commented how they look for a different key pieces of information in each days papers: Grocery ads on Tuesday, entertainment news on Thursday, church news on Friday, etc. One woman said that even the editorial opinions give her multiple pleasures. Sometimes she happily reads opinions she agrees with, but lots of times I like to get mad at it. I read it just to disagree with it. Also playing heavily in the ritual category was the attraction of crossword puzzles. Several readers mentioned that they eagerly look forward to the puzzle each day, and one even said she refused to buy a nearby metropolitan paper because the puzzle it runs was too difficult. Other newspaper items that readers said they read habitually were the advertisements (especially the Tuesday grocery ads), the recipes, the comics, the feature stories, a this date in history column and the letters to the editor. A

120 Almost without exception (and very often with a giggle or laugh), the respondents freely admitted to their newspaper habit. None, however, found anything negative in this and most said that if their own children and neighbors had a stronger newspaper habit, they would be better citizens. That the most frequently mentioned "soft" news content of the paper was the daily page of obituary notices is an indication of the way a newspaper moves its shared interactive functions out of the reader's house and into the community. We check the obits every night to see if our names are in there, said one man with humor in his voice. Then he seriously explained that in his community, it was important to know when a neighbor had died, both so one could express condolence and to discuss the death with other neighbors. Other readers noted that they read obituaries of people they have never known because they are interesting profiles of the people around them. That personal nature of the obituaries was once more prevalent throughout the paper, several elderly readers said. One 80-year-old woman noted that newspapers now run significantly less social news today than they did when she began reading as a youth. The lack of oldstyle society columns may be a reason newspaper readership has declined, she said. The personal importance of such news, she said, made her quickly come to rely on the newspaper as a social guide. While conceding

121 the topics of her youth may not be popular today, she said that finding similar means of attracting the emotions of readers should be more important to modern editors than it appears it is.

1998 Study: Discussion

There was no simple answer to the research question Can newspaper readers articulate why they feel a need to read? No, they could not spell out the definitive answer that almost any editor would pay to hear. But yes, they gave up important clues to the puzzle. The comments made by the participants in this research and those made in the 1945 New York study detail not one but many reasons why people like their newspaper when they have it and miss it when it is gone. The sheer number and variety of those comments demonstrate the tongueloosening effect of even a brief interruption of newspaper service, but also could be interpreted as confusion or an inability to define what is important about newspaper reading. By looking beyond the comments themselves and toward that unarticulated something, one begins, however, to see a trend that comfortably nests into the sociology roots of media dependency theory. The unifying function of the newspaper buried in the comments of the respondents was of social integration. Whether it was by providing them news of their neighbors, helping cope with the death of a friend or

122 simply telling them that tuna was on sale at the market, the newspaper made survival in their community much easier and more enjoyable, a function of the community-building ability of the press to which Janowitz (1952) alluded. This is also consistent with the premises of both the media dependency model (Ball-Rokeach 1998) and the uses and dependency model (Rubin and Windahl 1986). Despite this strong evidence of a social integration role for newspapers, one should not understate the hard-to-quantify comfort value of newspaper reading. This is closer to the notion of media use "outcomes" described in the uses and dependency model. (Rubin and Windahl 1986) The comment by one woman that watching TV is not like having a paper in the hands may be more salient than it first seems. As the respondents described their daily ritual of reading, there was obvious pleasure in their voices. The comfort provided in reading a familiar publication, with familiar content in their familiar easy chair had tremendous value to them. Part of the success of newspapers may be due to the mixture of legitimate news and community gossip that allows readers to fulfill their politically incorrect desire to be nosy in a way that is not only accepted, but lauded by society. This would mesh well with sociology-based theories of media dependency

123 If they just used the back-fence gossip pipeline, community news readers would be accused of cliquishness and scandal-mongering. But by reading about their neighbors in the paper, they appear legitimately informed. One respondent in this study put it in his own words: When you are reading, you are learning about the people you work with and grew up with, and people you dont even know. Certainly the ardor with which respondents defended their newspaper habit reinforced some of the premises Ball-Rokeach and others describe in media dependency theory, but the difficulty the respondents had tying their enjoyment of the paper to particular content is problematic. The newspapers local news and other content realistically could be replaced by other media. But time and again, the subjects of these interviews alluded to an indescribable something that made the newspaper itself important to them.

Preliminary Studies: Discussion

Both of these studies were as interesting for what they did not find as for what they did find. In the earlier study, the researcher sought to find a tangible factor either demographic or psychographic that would explain the varying level of intensity with which Americans read newspapers. There was none

124 to be found, however. A host of factors explained parts of the appeal of newspapers, but some hidden factor always seemed in the background. The second offered the advantage of free-wheeling discussion that makes qualitative research appealing. But, like most qualitative research, it was also limited in scope and its ability to generalize. Still, the respondents to the missing paper study again and again talked of their newspaper habit or their family rituals that included the newspaper. Is habit the mystery

"something" that drives a significant portion of readership? The fact that newspaper readers would so willingly describe their daily newsreading rituals was even more bewildering in light of the paucity of academic literature on the subject. Taken together, the two studies fairly scream for further research. Does the newspaper habit really exist, or is it just an urban myth? Can ritualized reading explain the anomalies in demographic readership predictors that point only to the old, the educated and the wealthy? Perhaps, as Berelson noted, the secret of media use lies not so much in the message, but in the desirability of reading. (Berelson 1949, Page 122)

125

CHAPTER VI

METHODOLOGY

This dissertation started with the premise that the appeal of newspapers is more than just the information they contain. But how can one winnow the desire for newspaper content from other factors that drive people to buy and read newspapers? The qualitative research in the 1998 preliminary study provided colloquial but unsubstantiated evidence that the habitual draw of newspapers is strong. The methodology of such a study is problematic, however. Disjointed quotations from multiple interviews can provide powerful pointers to truth, but they can also mask underlying factors that contribute to causality. Ideally, the question of why people read newspapers could be answered by a highly controlled experiment in which the demographics and attitudes of newspaper readers were measured and then the newspapers delivered to those readers were manipulated to vary content, delivery service and availability. In a free-market and open-information society, however, such an experiment is highly unlikely. Few readers are likely to agree to having their newspapers "censored," especially in the name of social science. And

126 few publishers are likely to risk alienating their customers by "playing" with content and delivery. Without the ability to test for habit in a controlled experiment, the logical next best methodology is to survey a scientifically valid sample and to analyze the data statistically to approximate the variable filtering of an experimental test though no survey can replicate that causality testing of an experiment. This chapter will first outline the research questions and hypotheses that guided the study. Next it will outline the procedures used to make such an inquiry of newspaper reader habits. Finally, it will explain the statistical tests that allowed the researcher to differentiate the effects of habit from other drivers of newspaper readership.

Research Questions

As previously stated, the intent of this research is to determine why people regularly read a newspaper. This dissertation, however, focuses on a single aspect of that broad question: the so-called "newspaper habit." The investigation of any suspected habit is burdened by a number of challenges. Is the behavior really a habit, or is it just repetition forced by outside factors, such as hunger or the daylight cycle? Is the habit strong enough to drive the behavior itself, or is it just an incidental result of

127 carrying out the behavior? Are there individual factors within the behavior that contribute to the habitual behavior? As these challenges are present in this dissertation, a logical structure to address them is the two-step process of research questions and hypotheses. The research question outline the philosophical questions to be addressed in the dissertation. The hypotheses establish premises, based on those questions, that can be tested through empirical means. This dissertation will work from the frame of four research questions and address eight hypotheses. Research Question 1: Does the "newspaper habit" really exist? As noted in Chapters 4 and 5, popular literature is filled with folk observations of ritual newspaper reading (Krim 1988; Lopez 1997; Saltzman 1995, p. 39) and many readers freely admit to the "newspaper habit" (Bentley 1998, p. 19). However, academic investigation to document whether this folk habit is a valid social phenomenon is missing from the literature. Research Question 2: Does the newspaper habit influence (drive) readership, or is it a reflection of readership driven by other factors? The mere existence of a reading habit is interesting, but it is of little consequence unless that habit has impact upon the readers or upon society. If some readers peruse their paper daily just out of habit, many of

128 the uses and gratifications assumptions that are based on the reader's use of newspaper content are called into question. Research Question 3: Do demographics, subscribership, specific content or other external factors contribute to the newspaper habit? If the newspaper habit does exist and it is a motivating factor in readership, it is natural to ask why. That question is best examined by first testing likely candidates for the answer. Both scholars and editors have pointed to demographics and to the various content elements of the newspaper as the stimuli that draw readers to newspapers. Research Question 4: Is the newspaper habit affected by the media dependency relationship of the reader? Media dependency theorists say that people turn to newspapers and other media not to fulfill some inner need, but as a tool to cope with the pressures of society (Ball-Rokeach 1998). If people are dependent upon newspapers for some social function, such as community identity, will they read the newspaper habitually?

Hypotheses

The following hypotheses quantify the issues raised in the research questions and allow them to be tested by statistical analysis: Hypothesis 1: Newspaper reading habit will positively influence people to read newspapers even when the effects of demographics and other traditional drivers of readership are controlled.

129

An index composed of observed repetitive behaviors associated with newspapers can be compared to measures of readership or importance of readership to test the "power" of the newspaper habit. This addresses Research Question 1. Research Question 2 requires that the power of habit to influence readers be tested against other more traditional drivers of readership, such as demographics. This can be accomplished through regression analysis, which allows dependent variables to be controlled. Hypothesis 2: The greater the readership of entertainmentbased content, the more a reader will display newspaper habit. Hypothesis 3: The greater the readership of local news-based content, the more a reader will display newspaper habit. Hypothesis 4: The greater the readership of obituaries, the more a reader will display newspaper habit. Hypothesis 5: The greater the readership of international news, the less a reader will display newspaper habit. Entertaining features, such as crossword puzzles and comics, along with local news and obituaries have high readership in newspapers (NAA 1998). But in the newspaper editorial culture, they are also considered reader attractants. International news is considered less attractive to readers because it does not reinforce community. Testing them against habit will determine whether their effect is ornamental or truly substantial. These hypotheses address Research Question 3. Hypothesis 7: The more readers use newspapers to reinforce their community identity, the more they will display newspaper habit.

130 Hypothesis 8: The more readers use newspapers to make their lives easier (utility), the more they will display newspaper habit. Community identity and ease of social navigation are primary independent variables of media dependency (Ball-Rokeach 1998). The uses and dependency model posits that habit and such goals are closely related.

Survey Overview

Data from two surveys will be used in this dissertation. The first study, conducted in 1998, was a statewide telephone omnibus survey that included a series of questions about newspaper use. The base survey for Study I was OASIS98, the first Oregon Omnibus Annual Social Indicators Survey, conducted by the Oregon Survey Research Laboratory from November 1998 to January 1999. The author participated in OASIS98 as part of a survey methodology class. The survey for Study II was the next edition of the Oregon Omnibus Annual Social Indicators Survey, OASIS99, conducted by the Oregon Survey Research Laboratory from November 1999 to December 1999. The author participated in OASIS99 as a fee-paying client. The first study, OASIS98, was conducted for a number of private clients and university researchers and asked Oregon adults 18 and older about a variety of subjects. These included gambling, smoking, camping,

131 religion, the environment, economics and (via the author's questions) newspaper reading habits. The survey employed a computer-derived random sample of all Oregon households, except those without telephones (4.5% according to the 1990 U.S. census). Calls were made through a computer-aided telephone interviewing system from 9 a.m. to 9 p.m. all days of the week, and up to 25 calls were made to each valid telephone number to avoid nonresponse bias. As a result, the overall response rate was 68.0%, as calculated under

the most rigorous guidelines of the Council of American Survey Research Organizations (CASRO). Interviews were completed with 805 persons, but the menu of questions was randomly divided so that 395 respondents were asked the newspaper questions. The margin of error for this sample in a 50-50 proportional split was 4.90 at the 95% confidence level. With the rigorous attention to random sampling and interviewer error exercised by the Oregon Survey Research Laboratory, the sample size was more than adequate to generalize to the Oregon population (Salant and Dillman 1994). Table 4 illustrates the relationship of sample size to population. The survey for Study II, OASIS99, was conducted along similar parameters. Like the first study it was conducted for a number of private clients and university researchers and asked Oregon adults 18 and older about a variety of subjects. These included Y2K concerns, religion,

132 population growth, the criminal justice system, plans for the Oregon State Library and (via the author's questions) newspaper reading habits. A computer-derived random sample of all Oregon households was used, excluding those without telephones (4.5% according to the 1990 U.S. census). Calls were made by trained interviewers through a computeraided telephone interviewing system from 9 a.m. to 9 p.m. all days of the week, and up to 25 calls were made to each valid telephone number to avoid nonresponse bias. The overall response rate was 57.58%, as

calculated under the Council of American Survey Research Organizations (CASRO) guidelines . Interviews were completed with 416 persons and all respondents were asked the same set of questions. The margin of error for this sample in a 50-50 proportional split was 4.79 at the 95% confidence level. As demonstrated in Table 4, with rigorous attention to random sampling and interviewer error, the sample size was more than adequate to generalize to the Oregon population (Salant and Dillman 1994). Demographic information on the respondents was compared to statewide population statistics to confirm that both the OASIS 98 and the OASIS99 sample compared favorably with the actual Oregon population.

133

TABLE 4. Valid Sample Size Needed for Various Populations (5% Sampling Error with 50/50 split) Population: 100 250 500 750 1,000 2,500 5,000 10,000 25,000 50,000 100,000 1,000,000 100,000,000
Adapted from Salant and Dillman (1994: p. 55)

Necessary Sample Size: 80 152 217 254 278 333 357 370 378 381 383 384 384

Study I Survey Questions

The OASIS98 survey asked 10 questions related to newspaper readership, which were clustered by theme. The first two questions concerned newspaper use: Now, thinking about the news, how important is it to you to read a newspaper regularly very important, somewhat important, or not important? Do you subscribe to a daily newspaper, a weekly newspaper or both? These two questions were meant to measure the intensity or level of involvement readers have with newspapers. The subscription question is self-evident, but may have less validity than the importance question. A

134 reported national growth trend in single copy (non-home delivered) sales of newspapers makes the subscription question problematic as a measure of readership. The Newspaper Association of America reported that 24% of weekday readers are single-copy buyers and 8% get their paper as a passalong from someone else. Only 68% of the regular readers in the NAA study were subscribers (NAA 1998). Using the NAA figures, threequarters of the nonsubscribers identified in the study are actually regular readers. Nonetheless, the wording of the subscription question gave some indication of reading frequency (weekly or daily) and to the type and number of publications to which the respondent subscribed. The importance question was a measure of where respondents placed newspapers in their lives and an attempt to circumnavigate the challenge of counting single-copy newspaper buyers. The next set of questions focused on the gratifications or consequences readers sought from newspapers Does reading the newspaper make you feel like part of your community? Do you read the newspaper for entertainment? Do you read the newspaper for relaxation? Do you read the newspaper to find information to make your life easier? Which of these reasons for reading a newspaper is most important to you?

135 The four questions focused on a variety of gratifications people might get from newspapers and could be used as measures of media dependency. The last three questions involved the physical activities that readers associate with newspaper use: Do you usually read the newspaper at the same time of day (when you read it)? Do you usually read the newspaper in the same place (when you read it)? Do you usually have the same thing to drink or eat while you are reading the newspaper? The three questions were designed to produce an index of newspaper ritual or habit. Time and location are intuitive indicators of habitual repetitive use. The third question was based on the folk observation of avid newspaper readers sipping coffee or eating donuts.

Study II Survey Questions

Study II was undertaken to expand on the original survey by using new criteria for newspaper readership and by probing for emotional indicators of reading habit. Specifically, the survey asked respondents to describe their emotional state when their reading routine is interrupted, and what portions of the newspaper content they read with regularity.

136 As a paid client of the Oregon Survey Research Laboratory, the author inserted 10 questions about newspaper use in OASIS99. The first question was a simple measure of readership: The next set of questions is about newspapers. In the average week, how often do you normally read a newspaper? This question, asking whether respondents normally read a newspaper every day, measures newspaper use differently than the first survey. Bowing to the national trend in single-copy (news rack) purchases of newspaper, it identifies frequent readers without requiring them to be subscribers. By employing a five-point scale, the question also examined reading rituals for various levels of infrequent readers as well as frequent readers. In addition, the scale allows for the inclusion of weekly newspapers in the review of habitual readership. Of the 90 newspapers in Oregon, 71 are weekly or semi-weekly publications (ONPA 1999). The next three questions sought to reaffirm the evidence of reader habit revealed in the first survey and were answered with simple yes/no responses: (When you read the newspaper) Do you usually read it at the same time of day? (When you read the newspaper) Do you usually read it at the same location? (When you read the newspaper) Do you usually read the pages of the newspaper in a particular order? The time and location questions were nearly identical to questions asked in the survey, with minor wording changes that reflected comments

137 raised during the Study II pretest. The last question in the group was substituted for the Study I question on food and drink to strengthen the ritual index. Reliability tests showed the food and drink question was the weakest of the original three. Question 5 was a quantitative followup to qualitative efforts by Berleson (Berelson 1949) and Bentley (Bentley 1998) to uncover the sociological "desirability" of reading. Which of the following statements best describe how you feel when circumstances prevent you from reading the newspaper as usual: Select all that apply to you: 1. 2. 3. 4. 5. 6. 7. I feel inconvenienced I feel uncomfortable I worry that I am missing out on something I feel that my routine has been disrupted I feel angry I feel confused I really dont miss it

The seven categories were taken from responses to interviews in the 1998 preliminary study. They were designed to measure the dominant emotion that habitual readers associate with their newspaper. The sixth question is another measure of the intensity of readership and habit: When circumstances prevent you from reading the newspaper as usual, what do you do with that days copy of the paper? Do you: 1. Keep it to read later 2. Tear out certain articles to read later

138 3. Promptly throw it away or recycle it 4. Give it to someone else 5. Other _________________ Answers one and two are indications that the readers feel they must examine the contents of the newspaper, even if they cannot do it on their regular schedule. Answers three and four indicate that the need to read the newspaper is transitory and time-related. Habitual readers who do not read a missed paper at a later date may have less need for the specific content than for the activity of reading. Questions 7, 8 and 9 recorded the popularity of content categories of the newspaper: What part of the newspaper is most important for you to read each time you get the paper? What part of the newspaper is second most important for you to read each time you get the paper? What part of the newspaper is third most important for you to read each time you get the paper? This was essentially one open-ended question asked in three waves. The three-tiered method was used because of the well-documented popularity of "news" or "local news" as a preferred content category (NAA 1998; Stone and Boudreau 1995); (Layton 1999). By recording second and third levels of popularity, the question quantified folk observations that people "only" get a paper to work the crossword puzzle (or read the comics, Dear Abby, etc.)

139 As in any open-ended question, the range of possible answers for the three questions was limitless. The author used his knowledge of newspaper content accumulated in 20 years in the newspaper industry to consolidate the answers into like groups. For instance, "comic strips," "funny papers," and "funnies" all became "comics." The answers to the open-ended questions were recoded into 32 categories:

140 Ads - Classified Ads - Coupons Ads - General Advice Columns Arts Business Calendar Comics Crime Crossword Current events Editorial Entertainment Headlines Horoscope International Letters to Editor Living Local news National news No Preference Obituaries Other Page 1 Politics Public Records Science, Health, & Tech Special news Sports TV Section Weather Whole Paper

141

142 Each content category was assigned a numerical value for use in statistical analysis, but the categories were not ranked nor considered interval data. The final question asked respondents about the term "newspaper habit': Some people say they have the newspaper habit. On a scale of 1 to 5, with 1 being no habit at all and 5 being a very strong habit, how strong of a habit is reading a newspaper for you? This question was reserved until after respondents were asked questions that made more subtle measures of habit or ritual. It provides two assessments. First, it allows the researcher to test the reliability of the ritual scale against the self-report of the respondents. Second, it gives an indication of the awareness respondents have to "newspaper habit" as a social term. The question employs a five-point, Likert-style scale that produced interval data. All questions were coded for a computer-aided telephone interviewing system by the staff of the Oregon Survey Research Laboratory. The accuracy of the coding was tested by employees of the lab, and the questions were pretested both by lab staff in an interviewertraining exercise and by a representative sample in a telephone pretest

143 that conformed to the operating standards of the Oregon Survey Research Laboratory.

The Study Area

Both Study I and Study II were conducted in Oregon. While Oregon's 3.32 million (U.S. Census Bureau 1998) people make up only 1.2% of the U.S. population, it is the 10th largest state in land mass, with 96,002 square miles (Oregon 2000). According to the U.S. Census Bureau, a somewhat lower proportion of Oregonians live in metropolitan areas than the national norm and slightly more are older than 65 than in the nation as a whole. Education, employment and household income approximate the national norm, although personal income per capita is somewhat lower. A significant difference between Oregon and the nation is the proportion of ethnic minorities in the population. Only 6.2% of Oregonians are non-white minorities, where the national figure is 17.2%. In addition, blacks comprise the largest minority group in the United States, but the largest minority bloc in Oregon is Asian and Pacific islanders (U.S. Census Bureau 1998). The details of how Oregon compares to the United States are listed in Table 5.

144

TABLE 5. Oregon Comparative Statistics


State POPULATION Resident population (1,000) 1999 Percent of ethnic minority (1998) White Hispanic Non-white Percent living inside metropolitan areas 1996 Percent under 18 years old 1998 Percent 65 years old and over 1998 EDUCATION Public elementary-secondary schools: Enrollment rate 1996 Full-time college enrollment, percent of total 1996 LABOR FORCE Percent of civilian population employed 1997 Average annual pay 1997 INCOME AND POVERTY Personal income per capita (constant 1992 dol.) 1998 Median household income (constant 1998 dol.) 1998 Percent of population below the poverty level 1998 Source: U.S. Census Bureau Rank U.S.

3,316 93.5 5.6 17.5 70.2 25.1 13.2

28

272,691 82.5 10.2 6.5

24 32 23

79.9 25.8 12.7

90.9 55.5

34 36

91.7 57.4

64.6 $28,420

27 22

63.8 $30,336

$21,975 $39,067 15.0

26 23 10

$23,436 $38,885 12.7

145

Limitations

This dissertation is subject to limitations of theory development and methodology. The theories of media use are in flux, with uses and gratifications theory competing with media dependency theory to explain essentially the same human conditions. The uses and dependency model offers hope for convergence, but also demonstrates that no one theory fully explains media use. This dissertation will further the convergence of uses and gratification with media dependency by exposing "habit" as a noncognitive factor in readership that has not previously fallen into the gratification category. The telephone survey methodology used in this research also has innate limitations. While telephone surveys have proven to be among the most reliable predictors of opinion and demographics, the proliferation of telemarketing and non-research sales techniques that emulate legitimate research have reduced the number of people willing to respond to a telephone poll. Also, telephone calls made in the daytime are more likely to connect to older, non-working adults. Those limitations are mitigated by the standard operating procedures of the Oregon Survey Research Laboratory. The laboratorys

146 computer assisted telephone interviewing (CATI) system varies the time of calls throughout the day and evening, and up to 25 calls are made to each valid telephone number to avoid non-response bias. In addition, telephone ownership is not universal in Oregon, so the segment of the population without home telephones will not be represented in this survey. Census data shows approximately 4.5% of Oregon households have no telephones. The Oregon Survey Research Laboratory has developed no means of reaching this minor portion of the population. Finally, the research for this dissertation was conducted in Oregon, a state with a smaller proportion of ethnic minorities than the average for the nation. Nevertheless, the literature does not show race as a strong predictor of readership. Studies such as the 1998 Newspaper Association of America media usage study (NAA 1998) show that blacks and other minorities read at a lower rate than whites, but that income and age are responsible for most of the difference. When those demographics were controlled through multiple regression in this author's analysis of OASIS 98, race was not a statistically significant factor in readership. Therefore, research on this admittedly restricted population should still have bearing on the national population.

147

CHAPTER VII

RESULTS OF STUDY I

Study I offered an initial look at habit or ritual in quantifiable terms, as opposed to the anecdotal terms of the 1998 initial study. The survey data from OASIS98 was well-suited to comparing the overall level of newspaper reading of respondents with their reported tendency toward habitual behavior. A set of questions in the survey also lent itself to analysis of the media dependency of respondents.

Variables

Two questions on the OASIS98 survey offered potential as dependent variables quantifying newspaper readership. The first simply asked if respondents subscribed to a daily newspaper, a weekly newspaper or no newspaper. The second asked How important is it to you to read a newspaper regularly very important, somewhat important, or not important? Of the 801 people who were asked about their newspaper subscriptions, 39.3% said they subscribed to a daily newspaper only, 10.2% said they subscribed to a weekly newspaper only, and 11.7% said they subscribed to both a daily and a weekly newspaper. No subscription

148 was reported by 38.5% of the respondents. Two respondents refused to answer the question. When the variable was recoded into a simple dichotomous measure of any kind of subscription, 491 respondents (61.3%) were subscribers and 308 (38.5%) were nonsubscribers. Of the 395 valid responses to the importance of newspaper reading question, 46.3% answered very important and 35.4% answered somewhat important. Just 17.7% 70 respondents said reading a newspaper regularly was not important to them. One respondent failed to answer and another answered dont know. A reported national growth trend in single copy (non-home delivered) sales of newspapers makes the subscription question problematic as a measure of readership. The Newspaper Association of America reported that 24% of weekday readers are single-copy buyers and 8% get their paper as a pass-along from someone else. Only 68% of the regular readers in the NAA study were subscribers (NAA 1998, p. 23). Using the NAA figures, a third of the nonsubscribers identified in the study should actually be regular readers. The NAA assertion was checked in the OASIS98 survey by comparing respondents to the subscription question with those of the newspaper importance question. A crosstabulation showed that 21.8% of the respondents who said they subscribed to no newspaper at all still said that reading a newspaper on a regular basis was very important to

149 them. An additional 41.5% of the non-subscribers said regularly reading a newspaper was somewhat important. While the first figure falls short of the 32% predicted by the NAA study, the fact that more than 60% of the nonsubscribers reported that reading is nevertheless important to them highlights the magnitude of the analytical problem of using subscriptions as a measure of readership. For this reason, the importance of reading variable was used as the initial dependent variable. However, the subscription information made a useful check variable. The survey question asking what types of newspapers the respondent subscribed to was recoded into a four-point scale variable, subscribership. As the sheer mass of multiple copies of

the newspaper in a household logically increases the chance of regular reading, the scale became incremental by ranking respondents from a low of non-subscription to a high of multiple subscriptions that arrive on multiple days of the week. Non-subscribers (38.5%) were assigned a value of 1, weekly subscribers (10.2%) were assigned a 2, daily subscribers (39.4%) were assigned a 3, and people who subscribe to both a daily and a weekly (11.7%) were assigned a 4. The function of the scale was to show an intensity of newspaper subscribership. Habitual reading of the newspaper was measured through three questions on reading logistics:

150 Do you usually read the newspaper at the same time of the day? Do you usually read the newspaper in the same place? Do you usually have the same thing to drink or eat while reading the newspaper? The three questions produced a low but acceptable Cronbachs alpha of .62 and were combined into an additive index, labeled news habit (Table 6).

TABLE 6. Newspaper Reading Habit Index Variables


OASIS98 Variable Question Percentage yes n

News 7 News 8 News 9 Cronbachs alpha Factor KMO

Do you usually read the newspaper at the same time of the day (when you read it)? Do you usually read the newspaper in the same place (when you read it)? Do you usually have the same thing to drink or eat while you are reading the newspaper? .62 .623

60.9% 73.2% 40.0%

325 325 325

An alternative combination of the news importance variables was obtained via factor analysis. The analysis produced a single factor with a Kaiser-Meyer-Olkin Measure of Sampling Adequacy of .623, well within acceptable norms. The Bartletts Test of Sphericity for the analysis was highly significant (p<.001). The factor analyses produced a newspaper habit variable with a range from -1.67 to 1.12.

151 The importance-of-reading question acted as a screening question for the habit questions and other questions on newspaper use. The small number of respondents who said newspaper reading was not important at all were not asked the other newspaper questions. This left 323 respondents in the valid sample. The survey also collected standard data on demographics, including sex, age, income, education and race. Race was not a significant variable for analysis, however, as the proportion of non-white residents in Oregon is less than 7% (Table 5 in Chapter 6). Political stance was measured through a question asking whether respondents considered themselves liberal (coded 1), conservative (coded 2) or moderate (coded 3) on economic issues. Sex was coded as 1 for males and 2 for females.

Method of Analysis

Crosstabulation, correlation, logistic regression and multiple regression were used in the analysis. As noted, regression analysis is a particularly appropriate technique for this type of research as it allows one to apply some of the data-sorting benefits of an experimental design to survey data. The regression equation not only allows one to plot a slope of increasing or decreasing effect, but to determine the impact on that effect if one controls various intervening factors. It does not, however, determine causality as is possible in experimental design.

152 Multiple regression was used when the dependent variable was continuous, as in the none-to-multiple papers variable of subscribership. Because the survey "screened" out the small number of respondents who said reading a newspaper was not important at all, the importance of newspaper reading variable became dichotomous. Dichotomous variables require the slightly less powerful technique of logistic regression. Unlike multiple regression, logistic regression cannot identify the portion of variance explained.

Findings

A simple look at the frequencies for newspaper habit indicated some habitual effect might be at work. Positive response to the three original questions about time, place and eating were very high, with more than 60% of respondents indicating they read at the same time of day and nearly three quarters saying they read in the same place (see Table 7). When the items were combined into the newspaper habit index, 62% of the respondents ranked in the moderate to high range. This supports Hypothesis 1: Newspaper reading habit will positively influence people to read newspapers even when the effects of demographics and other traditional drivers of readership are controlled.

153

TABLE 7. Importance of Newspaper Reading by Habit Level of Newspaper Reading Behavior


Level of News Habit Reading is somewhat Important Reading is very Important n

None Low Moderate High

57.1% 55.4% 35.0% 36.4%

42.9% 44.6% 65.0% 63.6%

56 65 100 99

TABLE 8. Means, Standard Deviations and Correlations for Variables


Read. import. Reading import. Habit Age Educ. Income Subscriber Political stance Sex (1=male) n Mean SD 1.00 .177** .129* -.050 .139** .475*** -.015 .027 393 2.28 .750 Habit .124* .139* .090 .257*** -.129* .005 321 1.75 1.08 Age .078* -.008 .218*** -.068 .083* 801 46.01 18.65 Educ. .171*** .036 -.010 .173*** 801 3.90 6.79 Income .273*** .021 -.083* 737 45.91 26.13 Subscr -.026 .033 799 2.24 1.09 Pol. stance -.047 772 2.24 .90 Sex 1.00 801 1.59 .53

* p<.05

**p<.01

***p<.001

154 Pearson correlation tests showed that the importance of newspaper reading has a statistically significant correlation with age (.129), income (.139), newspaper subscribership (.475) and the newspaper habit index (.177); but not education, sex or political stance (Table 8). The newspaper habit index was significantly correlated with middle age (.124), higher education (.139), increased subscribership (.257), conservative political stance (.129) and increased importance of newspaper reading (.177); but not with income or sex. The next round of analysis employed an elaboration model to check for effect. Simple crosstabulation tables were stacked to elaborate on the effect of adding variables to the bivariate relationship. The base table crosstabulating the importance of newspaper reading with habit reading behavior (see Table 8) showed that habit behavior generally increased with perceived importance of reading. When the crosstables were replicated for subsets of responses categorized by age, education, income and political stance, a similar pattern emerged, with notable exceptions. On the age table, (Figure 4) teens showed low levels of reading habit even if reading was very important, as did people in their 40s and people in their 70s.

155

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 20s 30s 40s 50s 60s 70s

Level of Ritual Newspaper Use None Low Moderate High

FIGURE 4. Age and Ritual Newspaper Use

Education also marched to its own drummer (Figure 5). People with high levels of education often had low levels of habit but place a high importance on reading. This indicated that more than one effect was acting on the importance of newspaper reading.

156

With that indication of effect, regression analysis was conducted on the variable for importance of newspaper reading, controlling for newspaper habit and the various demographics that other studies have shown to be strong indicators of readership. The importance of reading variable was recoded into a dichotomous variable for a logistic regression analysis. A five-step regression introduced a series of independent

100% 90% 80% 70% 60% 50% 40% 30% 20% 10% 0%

Level of Ritual Newspaper Use


None Low Moderate High

FIGURE 5. Education and Ritual Newspaper Use

157 variables. First newspaper habit was entered into the regression, then controls were added for age and education, income and sex, political stance and finally for newspaper subscribership. The regression analysis clearly showed that while the effect of habit on the perceived importance of reading a newspaper is small, it is significant, very definite and quite robust. As a single independent variable, habitual use of the newspaper was highly significant and generated a Chi-square of 9.26 (See Table 9). As expected, Model 2 showed that age and income were very powerful predictors of readership, increasing the Chi-square to 28.74. Nevertheless, habit remained significant even when age and income were controlled. This supports Hypothesis 1. Income and sex were added to the equation in Model 3, but were shown to be non-significant and reduced the B coefficient of the habit variable only minutely. Their addition to the equation, however, caused the Chi-square to become non-significant. When political stance was

entered in Model 4, it also proved to be non-significant, although it had a slightly stronger effect in reducing the B coefficient of habit. It was only when the subscribership variable was controlled in Model 5 that the equation changed radically and habit lost its statistical significance. However, this was due to the high correlation between

158 subscribership and importance of reading (r=.475). Subscribership also brought the Chi-square back up to 19.52 and high significance.

TABLE 9. B Coefficients for Logistic Regression of Importance of Regularly Reading a Newspaper. n=283
Independent variable Model 1 Model 2 Model 3 Model 4 Model 5

Newspaper habit Age Education Income Sex Political stance Subscribership Constant Chi square Degrees of freedom

.350** -325 9.26 1

.254* .035*** .231** -2.51*** 28.7 2

.253* .035*** .215* .003 .079 -2.72*** .411 2

.241* .035*** .214* .003 .058 -.112 -2.41** .58 1

.139 .029*** .200* -.002 -.071 -.076 .605*** -3.04*** 19.52 1

* p<.05 **p<.01 ***p<.001

Another asset of the logistic regression technique is that it allows one to forecast the probability that a variable will reach the highest value of the dependent value. By using the formula P=e B JX J / 1+ eB JX J , one can test the B coefficients of the variables to see how they apply in certain instances. When this formula is used to illustrate the effect of age and habit on the importance of newspaper reading, habit becomes much less of a mystery factor.

159 Figure 6 clearly shows that importance of reading a newspaper increases with both education and age. However, high levels of habitual use of the paper are associated with a high probability that the respondent feels that newspaper reading is very important whether educational levels are high or low. A probability chart with education in the X axis and age in the interaction produces a very similar plot. Again, this supports Hypothesis 1. To further test the hypothesis that habit increases readership, a corresponding regression equation was calculated with the four-level intensity of subscribership item as the dependent variable. In addition

to providing another measure of readership, the continuous nature of the subscribership variable allowed the use of multiple regression analysis.As Table 10 shows, the effect of habit on the subscribership measurement was even stronger than it was on the importance of reading measurement. The bivariate regression analysis in Model 1 had an R2 of .061, indicating it explained more than 6% of the variance. Habit was highly significant in Model 1. As in the earlier analysis, the addition of age and education to the equation in Model 2 had a profound effect. The R2 more than doubled the B coefficient to .125, and the B coefficient of habit dropped somewhat to .195. However, habit remained highly significant.

160

90% 80% 70% 60% 50% 40% 30% 20% 10% 0% 20 30 40 Age
Hi ritual/Hi ed Lo ritual/Hi ed Hi ritual/Lo ed Lo ritual/Lo ed

50

60

FIGURE 6. Importance of Newspaper Reading, by Age

161 In a change from the earlier analysis, Models 3 and 4 showed that controlling for education was not significant, but controlling for income was. This may be a reflection of the overtly monetary nature of subscribing to a newspaper as opposed to the more emotional nature of stating that reading a newspaper is personally important. Habitual use of the newspaper remained significant in Model 5, when the importance of reading was controlled. This analysis reaffirmed support for Hypothesis 1, the primary hypothesis about habit. As Table 10 shows, the effect of habit on the subscribership measurement was even stronger than it was on the importance of reading measurement. The bivariate regression analysis in Model 1 had an R2 of .061, indicating it explained more than 6% of the variance. Habit was highly significant in Model 1. As in the earlier analysis, the addition of age and education to the equation in Model 2 had a profound effect. The R2 more than doubled the B coefficient to .125, and the B coefficient of habit dropped somewhat to .195. However, habit remained highly significant. In a change from the earlier analysis, Models 3 and 4 showed that controlling for education was not significant, but controlling for income was. This may be a reflection of the overtly monetary nature of subscribing to a newspaper as opposed to the more emotional nature of stating that reading a newspaper is personally important. Habitual use of the newspaper remained significant in Model

162 5, when the importance of reading was controlled. This analysis reaffirmed support for Hypothesis 1, the primary hypothesis about habit. Finally, responses to the four "dependency" questions were compared. The questions asked: 1. "Does reading the newspaper make you feel like part of your community?" 2. "Do you read the newspaper for entertainment" 3. "Do you read the newspaper for relaxation?" 4. "Do you read the newspaper to find information to make your life easier?"

TABLE 10. Unstandardized Ordinary Least Square Coefficients for Regression of Intensity of Newspaper Subscription. Independent variable Model 1 Model 2 Model 3 .191*** .013*** .048 .008** .021 .166*** 8.17*** Model 4 .185*** .013*** .046 .008** .090 -.064 .169*** 6.91*** Model 5 .155** .008* .020 .007** .090 -.051 .542*** .191*** 9.36***

Newspaper habit .243*** .195*** Age .013*** Education .088* Income Sex Political stance Importance of reading a newspaper R2 (Portion of variance .061*** .125*** explained) F 18.46** 13.47** * * p<.05 **p<.01 ***p<.001 n=286

163 Figure 7 illustrates the comparison of those respondents who answered "yes" to the above questions. The analysis shows the differences between people whose answers to the newspaper habit index showed they had no habit, those who had a low level of newspaper habit, those who had a moderate level of newspaper habit and those who had a high level of newspaper habit. For comparison, the "yes" responses of the total survey sample are also shown. The analysis shows interesting differences among the populations of readers. People who read to build a sense of community are more likely to have a high level of newspaper habit than others. This supports Hypothesis 6 (The more readers use newspapers to reinforce their community identity, the more they will display newspaper habit). The data show people who read for entertainment are more likely to have a low level of newspaper habit. This is counter to Hypothesis 3 (The greater the readership of local news-based content, the more a reader will display newspaper habit) . But people who read for relaxation have a high level of habit, which supports Hypothesis 3. Reading for information to make life easier is popular with everyone, but it is especially popular with people with no newspaper habit at all. This is directly opposite of the prediction in Hypothesis 7 (The more readers use newspapers to make their lives easier (utility), the more they will display newspaper habit).

164

60%

50%

40%

30%

20%

10%

0% Community Entertainment Relaxation Information

Level of Newspaper Habit


None Low Moderate High

FIGURE 7. Habit and the Most Important Factor in Reading

Nevertheless, the observations cannot be taken as full support or denial of the hypotheses. A Pearson correlation test of the four dependency questions with the newspaper habit index failed to produce statistical significance. For this reason, the comparisons are inconclusive.

165 The statistical evidence that habit increases newspaper readership was novel enough in the literature to raise questions about the procedures used. For this reason, a second study was launched to verify the data. This also presented the opportunity to fine-tune some of the questions and to add new questions to attempt a more definitive test of Hypothesis 7, the hypothesis on utility. The results of this second study are reported in Chapter 8.

166

CHAPTER VIII

RESULTS OF STUDY II

Study II was undertaken with two major goals. First, it offered an opportunity to verify the finding in Study I that habit has a distinct influence on newspaper readership. While habit and newspapers are discussed in the literature, no statistical proof of the effect is extant. A retest of the procedure used in Study I was therefore prudent. Second, Study I left open the question of whether the enhancement of newspaper readership by habit was itself influenced by a preference for particular parts of the paper. Respondents to the 1998 qualitative study had mentioned comics, crosswords and obituaries as influences on their desire to read the newspaper.

Variables

The OASIS98 used a self-reported index of the importance of reading a newspaper and subscription to daily and weekly newspaper to survey as dependent variables quantifying newspaper readership. OASIS99, upon which Study II was based simply measured the frequency at which respondents "normally" read a newspaper. Similar indices have been used

167 to measure newspaper readership for several years (NAB 1978; Poindexter 1979; Westley and Severin 1964). Of the 419 people who responded to the question "In the average week, how often do you normally read a newspaper?," 14.7% said they never read and 3.8% said they seldom read. A much larger proportion read at least one day a week: 27.3% said they read 1-3 days a week, 11.2% said they read 4-6 days a week and 42.5% said they normally read a newspaper daily (Table 11). This compares with the Newspaper Association of America study that showed 25% of Americans were "nonreaders," 24% were "occasional readers" and 51% were "frequent readers." That same study reported that 59% of the respondents read a daily newspaper "yesterday" and 67% had read the previous Sunday's paper (NAA 1998, p. 18).

TABLE 11. Frequencies for "In the Average Week, How Often Do You Normally Read a Newspaper?"
Value Label Value Frequency Percent Cum Percent

Never Seldom 1-3 days a week 4-6 days a week Every day

1 2 3 4 5

62 16 115 47 179

14.7 3.8 27.3 11.2 42.5

14.8 18.6 46.1 57.3 100

168 Habitual reading of newspapers was measured through two variables. The first was a "habit index" very similar to that developed in Study I. Participants were asked three questions: 1. (When you read the newspaper) Do you usually read it at the same time of day? 2. (When you read the newspaper) Do you usually read it at the same location? 3. (When you read the newspaper) Do you usually read the pages of the newspaper in a particular order? Questions 1 and 2 were nearly identical to the habit index questions asked in Study 1. Question 3 substituted for a question on the choice of the same food or beverage to consume while reading the paper, as this question produced the lowest correlation coefficient in the reliability test that produced the index. Refusals and invalid answers reduced the size of the sample answering all three questions to 357. Table 12 shows how responses to the three questions were combined into a single index. The first question, asking if respondents usually read at the same time of day, produced a 60.5% positive response. Location was even more "habitual," as 79% of the respondents said they read in the same place. The new question on order of pages also drew a strong positive response, with 61.1% answering yes. When the results were combined into an additive index (labeled "habit") via the SPSS statistical program, they produced a Cronbach's alpha of .59.

169 TABLE 12. Study 2 Newspaper Reading Habit Index Variables


OASIS99 Variable Question Percentage yes n

Paper2 Paper3 Paper4

When you read the newspaper, do you usually read it at the same time of the day? When you read the newspaper, do you usually read it at the same location? When you read the newspaper, do you usually read the pages of the newspaper in a particular order? .59 .61

60.5% 79.0% 61.1.0%

357 357 357

Cronbachs alpha Factor KMO

Although that is an acceptable level of reliability, it was both on the low end of the acceptable scale and was lower than the Cronbach's alpha of Study 1. A second test using factor analysis, therefore, was used to confirm the validity of the index. The analysis produced a single factor with a Kaiser-Meyer-Olkin Measure of Sampling Adequacy of .605, within acceptable norms (UTSA 2000). The Bartletts Test of Sphericity for the analysis was highly significant (p<.001). The level of habit in respondents was also self-measured through a direct question (labeled "My habit"): Some people say they have the newspaper habit. On a scale of 1 to 5, with 1 being no habit at all and 5 being a very strong habit, how strong of a habit is reading a newspaper for you?

170 Like Study I, the OASIS99 data in Study 2 included standard demographic information. Two additions added to the omnibus survey by other researchers also proved useful, as they had "face value" indications of regular or habitual lifestyle characteristics. One was a question asking whether the respondents were registered voters, indicating a willingness to participate in periodic political activities. The second was a question asking how important religion was in their life. As most religions encourage weekly if not daily observances and are the basis for most definitions of ritual (Bell 1992; Carter 1999; Durkheim 1915), the results were added to the analysis of newspaper habit.

Method of Analysis

Crosstabulation, correlation and simple comparison of means was used in the initial evaluation of the data. Because the reading frequency variable was continuous and asked of all respondents, multiple regression was used to test for the effect of habit instead of the less-powerful technique of logistic regression (used in Study 1). As noted previously, regression analysis allows researchers to filter the effects of multiple independent variables on a dependent variable.

171 Findings

The data fulfilled Study II's initial goal of verifying the effect of newspaper habit to a pleasantly high degree. As in Study I, the means of questions used to create the newspaper habit index showed that the majority of readers exhibited some sort of habitual behavior when reading the newspaper (Table 13), supporting Hypothesis 1 ( Newspaper reading habit will positively influence people to read newspapers even when the effects of demographics and other traditional drivers of readership are controlled): TABLE 13. Frequencies for Media Habit Index Value Label No habit Some habit Average habit Strong habit Very strong habit n=420 Value 1 2 3 4 5 Frequency 88 70 90 56 49 Percent 24.9 19.8 25.4 15.8 13.8

A correlation analysis also verified the findings of Study I (Table 14). The reading frequency variable was significantly correlated with the habit index, age, education, income, sex, voter registration and the self-reported newspaper habit measure. The Pearson correlations for several measures were within a few thousandths of the measures in Study 1. The habit index was statistically significant in its correlation with reading frequency, age, voter registration and the self-reported habit measure, but was not

172 significant for education, income, religion, sex or whether the respondent lived in an urban or rural community. The self reported measure of newspaper habit, however, correlated with statistical significance with the frequency of reading, the habit index, age, income, and voter registration. TABLE 14. Means, Standard Deviations and Correlations for Study 2 Variables
Freq Read
Freq. of Reading Habit Age Educ. Income

Habit 1.00 -.304 *** -101. .003 -.005 -.023 -.020 .157 ** .386 *** 357 2.01 1.03

Age 1.00 -.084 -.001 -.073 -.036 -.069 -.200 *** .318 *** 409 53.64 17.68

Educ. 1.00 .211 *** -.096 -.124 * -.069 .199 *** .091 409 3.24 1.47

Inc. 1.00 -.039 -146 ** .092 .079

Relig 1.00 .190 *** .027 -.020

Sex 1.00 -.019 -.094

Urb/ Rur 1.00 -.128 * -.050 409 4.77 5.25

Vote 1.00

"My" habit

1.00 -.386 *** .241 *** -.241 *** 127 *

Import. of -.000 Relig. Sex (1=M, 2=F) Urban or rural Reg. Voter (0=No, 1=Yes) "My" habit n Mean SD

114 ** .094 -.244 *** 51 *** 419 2.37 1.43

.117* 407 4.27 1.95

.040 409 2.75 1.09

-.021 409 1.55 .50

.211 *** 409 .87 .33

1.00 354 2.75 1.39

* p<.05 **p<.01 ***p<.001\

173 Of greater interest, however, is the regression analysis of the newspaper reading frequency variable with the two measurements of habit and a variety of demographic measurements (Table 15). As in Study 1, the regression analysis revealed a small but very robust and statistically significant effect of habit on newspaper reading. The data for Study I were regressed in six steps, producing six models to control for the secondary effects of demographic information. Model 1 examined frequency of reading by the effect of habit alone. The test produced a highly significant B-coefficient of .370 (p<.001) and an R-square of .147. The R-square, when converted to a percentage, represents the portion of the effect explained by the model. In this case, it indicates that nearly 15% of the frequency of newspaper reading is explained by the habit index. Model 2 adds the demographic factors of age and education that earlier researchers (Stone 1987) have credited with the primary influence on readership. Indeed, the power of their influence is borne out by the regression analysis, as each variable is highly significant (p<.001) and the R-square (or portion of explained effect) nearly doubles to .264 or 26.4%. Notably, however, habit remains highly significant (p<.001)

despite the addition of age and education. This supports Hypothesis 1.

174

TABLE 15. Unstandardized Ordinary Least Square Coefficients for Regression of Importance of Frequency of Reading a Newspaper. n=347 I ndependent variable Newspaper habit Age Education Income Sex Registered voter Importance of religion Urban or rural Self-reported newspaper habit Constant R2 (Portion of
variance explained) * p<.05 **p<.01 ***p<.001

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 .370*** .285*** .280*** .266*** -.016*** .016*** -.016*** .126*** .095** .088** .067** .064** -.221* -.220* .376* .042 .262*** -.016*** .083** .063** -.216* -.369* .044 -.044* .160*** -.012*** .071** .047* -.221* .249 .036 -.040* .246*** 3.50*** .410

3.36*** 3.97*** 4.15*** 3.71*** 3.96** .147 .264 .293 .308 .319

175 Model 3 adds two more potent demographics, income and sex. Again, both variables were statistically significant but did not eliminate the significance of habit. The negative B-coefficient of sex indicates that the frequency of reading effect was stronger among males than females a trend identified in national readership research (NAA 1998). This second set of demographics had only a small impact on the overall effect, raising the R-square to just .293 or 29.3%. In Model 4, the two intuitively habitual questions voter registration and importance of religion added to the survey by other researchers were placed in the regression as independent variables. Only voter registration was statistically significant and neither diminished the significance of habit. The increase of explained variance (R-square) was quite small, to 30.8%. Model 5 added just one more variable, a measure of how urban or rural was the hometown of the respondent. The variable was significant (p<.05) in the negative direction, which indicates that people who live in more urban areas tend to read newspapers more frequently. As urban newspapers are usually published daily while many rural newspapers are weekly, this is a logical finding. The urban/rural variable did not, however, change the significance of the newspaper habit variable (p<.001). The addition of the urban/rural variable raised the R-square to 31.9%.

176 The regression analysis was completed with Model 6, which added the self-reported newspaper habit variable. The variable was highly significant (p<.001) and raised the explained variance to 41% (R2 -=.410) but did not eliminated the significance of the habit . A logical explanation of this is that the two habit variables measured approximately the same effects, so there was little more variance to be explained by the addition of the second habit variable. The robustness of the habit index variable when the secondary effects of other variables are controlled further reinforces support for Hypothis 1 . To test Hypotheses 2-7, the frequencies of the question "What part of the newspaper is most important for you to read each time you get the paper?" were analyzed. The open-ended question was asked in three waves ("what part is most important, what part is second most important, what part is third most important ...") to mitigate the overwhelming popularity of news. (NAA 1998) The strength of that popularity was borne out by an examination of the rankings people gave the various sections in each of the three waves (Tables 16-18). The significance of the rankings was tested via Spearman correlations among the categories and between the respondent groups. All correlations were highly significant (p<.001). "Page 1" was the overwhelming (32.5%) preference in the first round of choices, followed by

177 local news (13.6%), sports (7.9%), business page (6.2%) and the classified ads (4.5%). Local news rose to the top position in the second round and sports in the third round, though the "hard news" categories tended to stay in the highest tier of preferences in all three rounds. Nevertheless, several "soft" categories, such as the living page, the editorial page and general advertising rose in importance in the second and third rounds of choice. The SPSS statistical program provides a means of simplifying such an array of answers via its "multiple responses" command. The command creates a new variable that consolidates the three sets of answers in a form that can be used for crosstabulation and frequency analysis. The effect of this consolidation can be seen by comparing Table 16 and the "All Respondents" column of Table 17. The data for Table 17 were generated via the SPSS multiple responses command. Just as in the "first choice" rankings of Table 16, "Page 1," "local news," and "sports" rank 1, 2 and 3 in the consolidated data of Table 17. But the fourth place ranking changes from "business" to "living" when the three waves of choices are consolidated. "Comics" slips down a ranking point, from sixth to seventh. More changes in the ranking lineup are evident as one looks lower on the two lists.

178 TABLE 16. First, Second and Third Choice of Most important Section or Page to Read
First Choice
Value Label Rank Freq Pct.

Second Choice
Rank Freq Pct.

Third Choice
Rank Freq Pct.

Page 1 Local news Sports Business Ads - Classified Comics Headlines Living Editorial Current events International No Preference Science, Health, Whole Paper Letters to Editor Obituaries Crossword Weather Ads - General Crime Public Records National news Special news Ads - Classified TV Section Ads - Coupons Horoscope Arts Entertainment Calendar/events Advice Columns Other
* = Top 5 choice

* 1 115 32.5 * 2 48 13.6 * 3 28 7.9 * 4 22 6.2 * 5 16 4.5 6 16 4.5 7 14 4 8 12 3.4 9 12 3.4 10 11 3.1 11 10 2.8 12 8 2.3 13 6 1.7 14 6 1.7 15 5 1.4 16 4 1.1 17 4 1.1 18 3 0.8 19 3 0.8 20 2 0.6 21 2 0.6 22 1 0.3 23 1 0.3 24 1 0.3 25 1 0.3 26 1 0.3 27 1 0.3 28 1 0.3 29 0 0 30 0 0 31 0 0 32 0 0
Ttl. 354 100.1

*2 *1 *3 6 *5 7 16 *4 11 17 12 8 20 22 23 13 21 14 9 27 28 10 24 29 31 30 32 25 18 26 15 19
494

52 67 45 17 18 16 4 36 9 4 9 13 3 2 2 8 3 5 11 0 0 11 1 0 0 0 0 1 4 1 5 4
347

14.8 19.1 12.8 4.8 5.1 4.6 1.1 10.3 2.6 1.1 2.6 3.7 0.9 0.6 0.6 2.3 0.9 1.4 3.1 0 0 3.1 0.3 0 0 0 0 0.3 1.1 0.3 1.4 1.1
98.9

*5 *2 *1 8 *4 7 18 6 9 19 13 *3 27 12 29 10 28 11 16 17 30 14 24 21 32 31 26 20 15 25 22 23

27 7.8 37 10.8 39 11.3 22 6.4 29 8.4 25 7.3 4 1.2 27 7.8 16 4.7 4 1.2 8 2.3 34 9.9 1 0.3 9 2.6 0 0 9 2.6 1 0.3 9 2.6 6 1.7 6 1.7 0 0 7 2 2 0.6 3 0.9 0 0 0 0 2 0.6 4 1.2 7 2 2 0.6 2 0.6 2 0.6
99.4

490 342

Percentages do not total 100 due to rounding.

179

For ease of comparison, the consolidated list is preferable and was used to analyze the content preferences of newspaper consumers based on their position in the newspaper habit index. Table 17 shows the rankings respondents gave to 32 newspaper sections or pages. The rankings are listed for all respondents, for those who indicated "no habit" on the index, for those who indicated "low habit," for those indicated "moderate habit" and for those who indicated "high habit." Table 18 shows the similar rankings respondents gave to 32 newspaper sections or pages, but this time is split by the levels of the selfreported newspaper habit variable. The rankings are listed for all

respondents, for those who claimed "no habit," for those who indicated "some habit," for those of "average habit" and for those who claimed "strong habit" and for those who said they have a "very strong" newspaper habit. It is worth repeating that the significance of the categorical data was tested by conducting Spearman correlations between the respondent groups of the categorized lists. All correlations were significant (p<.001). The tables provide several interesting observations. Classified advertising on Table 17 holds a much higher rank for respondents with no habit (4) than with respondents with high habit (8). The difference is even more dramatic on the self-reported habit scale, with readers claiming

180 no habit ranking classified ads third, but readers who said they have a very strong newspaper habit rank classified ads 14th. Readers with high habit, however, give more importance to "comics" than do no-habit readers (6 vs. 9 on Table 17, 4 vs. 8 on Table 18). This could be an indication that no-habit readers consume the newspaper in a utilitarian manner (shopping) while high-habit readers seek specific content. As was earlier noted, this rejects Hypothesis 7. To test this contention and similar contentions in a methodical manner, the relative rankings of various types of content by respondents with no habit and by those with high habit were compared. Hypothesis 2 said that high interest in entertainment-based content corresponds with high newspaper habit. For this, the rankings of the "living," "comics," "entertainment," "crossword," "advice columns," "arts" and "horoscope" and "TV" pages were analyzed. Respondents with high newspaper habit scores ranked all but the "entertainment" page higher. While this seems out of character, Hollywood or entertainment news is often treated as a genre of its own and relegated to "tabloid" journalism. It may also be age-specific rock music reviews for teens, classical music reviews for older readers. The higher rankings for personally entertaining pages supports Hypothesis 2.

181 TABLE 17. Rankings of Most Important Section, by Level of Newspaper Reading Habit Index
Section or Page Most Important to Read All Respondents No Habit Low Habit Moderate Habit High Habit

Page 1 Local news Sports Living Ads - Classified Business Comics No Preference Editorial International Headlines Obituaries Ads - General Current events National news Weather Whole Paper Entertainment Science, Health, Tech Crime Crossword Letters to Editor Advice Columns Arts Other Special news Politics Horoscope Calendar/events Public Records TV Section Ads - Coupons
Between-group correlations p<.001

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

1 2 4 6 3 7 9 5 8 10 12 14 15 11 24 13 17 18 16 25 19 20 26 27 28 21 29 30 22 23 31 32

1 2 3 6 7 8 5 4 10 9 17 22 14 11 12 13 16 15 21 18 30 24 32 20 25 19 26 31 23 27 28 29

1 2 3 5 4 7 8 6 10 12 14 11 9 13 19 18 21 15 20 16 23 24 17 22 28 31 29 25 26 27 32 30

1 2 3 4 8 5 6 9 7 13 10 12 20 16 11 15 14 23 18 25 17 21 24 26 19 28 22 27 30 31 29 32

TABLE 18. Rankings of Most Important Section by Level of Self-Reported Newspaper Habit

182
Section or Page Most Important to Read All Resps. No Habit Some Habit Average Habit Strong Habit Very Strong Habit

Page 1 Local news Sports Living Ads - Classified Business Comics No Preference Editorial International Headlines Obituaries Ads - General Current events National news Weather Whole Paper Entertainment Science, Health, or Technology Crime Crossword Letters to Editor Advice Columns Arts Other Special news Politics Horoscope Calendar/events Public Records TV Section Ads - Coupons Hypothesis 3

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 said the

1 2 5 6 3 7 8 4 11 15 14 20 10 9 16 12 13 24

1 2 3 5 4 6 7 8 9 12 11 13 14 10 24 22 23 16

1 2 3 4 7 6 5 8 9 11 15 12 10 18 13 14 21 19

2 1 3 4 10 5 8 9 11 6 13 15 24 20 7 12 22 14

1 2 3 6 14 5 4 13 7 12 8 9 21 18 11 16 10 17

17 19 16 16 23 18 15 22 25 22 25 17 23 17 19 21 20 24 23 20 26 18 20 21 26 27 27 17 19 25 22 25 30 30 15 19 28 25 26 27 28 29 26 18 24 29 21 29 29 30 23 26 31 31 31 30 30 27 27 28 31 31 32 32 32 32 32 28 28 29 appeal of local news-based content is related to

high habit. For this, the local news, living, current events, and crime pages were reviewed. The results were inconclusive. The "local news"

183 pages held their second-place rank no matter what the habit score of the respondent was. "Living" gained rank, "current events" lost rank and "crime" stayed even. This may be a reflection of the previously mentioned overwhelming popularity of "news." Hypothesis 3 was not supported. Hypothesis 4 focused on obituaries, which were extremely popular for respondents to the 1998 qualitative preliminary study. Indeed, obituaries ranked well above the midpoint for respondents of all habit levels in Study 2. Respondents with no habit ranked obituaries 14.

Respondents with moderate habit ranked obituaries 11. Respondents with high habit ranked obituaries 12. Hypothesis 4 was supported. With the assumption that international news is less attractive because it does less to reinforce community than do local stories, Hypothesis 5 said that high readership of international news will indicate low habit. Indeed, the rankings given "international pages" fell as habit increased. Respondents with no habit ranked international news 10 and those with high habit ranked international news 13. Hypothesis 5 was supported, although only marginally because of the small shifts in rankings. Hypothesis 6, that readers use newspapers to reinforce their community identity, was supported in Study 1. Marginal support was added, however, by the rankings of "editorial pages." Respondents with no habit ranked editorial pages 8, while those with high habit ranked them 7.

184 However, that order was reversed for "letters to the editor," with no-habit readers ranking them 20 and high-habit readers ranking them 21. Hypothesis 7 said that readers who use the newspaper to make their lives easier (utility) will have higher levels of newspaper habit. Although one might argue that the high emphasis on news by readers with elevated levels of habit and their high rankings of advice columns are indicative of a preference for utility, the low rankings given such as current events, weather, public records and calendars appear to reject Hypothesis 7 and support the counter idea that readers with lower habit levels seek utility from the newspaper. The anomaly is the television section, which gets a high ranking from readers with high habit. That ranking, however, may be related to their preference for entertainment content. Two other questions in the survey were less tests of the relative strength of the newspaper habit than they were an attempt to identify "markers" by which future researchers could identify habitual newspaper readers. The first of these asked respondents to identify how they felt when circumstances prevented them from reading the newspaper as usual (Figure 8) . The second asked what they do with a newspaper that they have not had a chance to read as usual (Figure 9). Many (45.5%) of the surveyed newspaper consumers said they did not miss the paper when circumstances prevented them from reading it.

185 Figure 6 demonstrates, however, substantial differences among those who do miss reading their newspaper. Respondents with a high level of newspaper habit as measured on the habit index were clearly more upset about missing the paper than were other respondents. Where only 4.7% of those persons in the "no habit" category said they felt "inconvenienced" by the missing paper, 29.7% of those with "high habit" felt inconvenienced. Another 13.5% of the high habit readers felt they were "missing something" and 6.8% felt their routine was disturbed. This could indicate both a liability and an opportunity for newspaper publishers. The trouble spot arises from the possible personal affront that delayed or misrouted newspapers can represent to "inconvenienced" high-habit readers. In addition, the high number of people who said they may be "missing something" are prime targets for alternative sources of information, such as broadcast or the Internet.

186

60% 50% 40% 30% 20% 10% 0% No habit Inconvenienced Routine disturbed Low habit Moderate habit High habit

Uncomfortable Angry

Missing something Don't miss it

FIGURE 8. How Do You Feel When You Are Prevented from Reading the Newspaper?

187

60% 50% 40% 30% 20% 10% 0% No habit Low habit Moderate habit Tear out articles Give it away Don't know High habit

Keep it Throw itaway/recycle it Other

FIGURE 9. Disposal: What Do You Do with an Unread Paper?

188 The opportunity for publishers comes in responding to the importance of receiving the newspaper for high-habit readers. Home delivery, long-term subscriptions and on-time delivery guarantees should be attractive to these readers, who would be inconvenienced by a "missed" delivery. The analysis of newspaper disposal (Figure 9) also offers good news to publishers and insight into the newspaper habit. When faced with a newspaper that they were prevented from reading for some reason, 43.2% of the respondents said they would simply keep it for later consumption. The trend to keep the old paper for later use held across the habit spectrum, with even 41.9% of those in the "no habit" category saying they would keep the paper. Of special interest is the observation that persons with no and low newspaper habit were among the most common respondents to say they "tear out certain articles to read later." With additional research to

identify the type and format of articles most frequently "clipped," newspapers could use this tendency to appeal directly to lower-habit readers. The over-arching contribution of Study 1 is its strong evidence of a quantifiable "newspaper habit" phenomenon that can be identified via an index of behaviors. How that phenomenon can contribute to communication scholarship will be discussed in Chapter 9.

189

190

CHAPTER IX

DISCUSSION

The ritual consumption of print media specifically, the newspaper habit has import in both the theoretical and applied arenas. The ubiquitous nature of newspapers in America means that any habit associated with them is a habit of a whole nation. The newspaper habit also may play a role in the dissemination of public policy information, and it could affect the economic viability of publications. Every researcher hopes to discover in his or her data a "landmark" revelation that will clearly show the whole scholarly world a new truth. Realistically, this is seldom the case. This dissertation instead offers a small number of "ah-ha" findings, a greater number of eyebrow-raising implications and more than few new puzzles that can only be addressed by further research. This chapter will attempt to put those modest accomplishments into a digestible perspective. It will first review the face-value findings of the hypotheses tests, then review the implications of the results and finally offer recommendations to researchers for further study and to practitioners about how the results may be used by the newspaper industry.

191 The data presented here support the assertion that habit has a substantial impact on the consumption of newspapers. By association,

one can speculate that ritual or habit are among the hidden factors that have kept the newspaper popular with Americans despite the introduction of faster, more colorful and less expensive electronic alternatives. Nevertheless, a review of the seven hypotheses upon which the data was based may be useful before discussing the general implications of this study. A summary of the hypotheses is in Table 19.

Review of the Hypotheses

Hypothesis 1 was the main assertion upon which this dissertation was based: Newspaper reading habit will positively influence people to read newspapers even when the effects of demographics and other traditional drivers of readership are controlled. This hypothesis was supported by several tests conducted during the research. This is not surprising, as the folklore of the newspaper reading habit is strong. However, there was little empirical data to verify the folk observations. The simplest support came from an analysis of the means in both Study I and Study II. In both cases, a majority of respondents said they read their newspapers with some sort of habitual behavior: They usually read at the same time of day, read it in the same location, enjoyed the

192 same refreshments (like a cup of coffee) with it or read the sections of the paper in the same order. This provided the basis for the statistical indices of newspaper habit used in the data analyses of Study I and Study II. This should not be surprising. Philosophers from Aristotle (Cooper 1932) to James (1914) cite the penchant of humans to conveniently relegate daily tasks to habit, conserving the energy of concentration for more critical tasks. While the simple means of the survey responses supported the observation that habit was present, it took additional tests to link that habit with newspaper readership patterns. This was done by correlating the habit index both by the readers' judgment of how important regularly reading a newspaper is to them, and by testing it against their self-reports on the level of their "newspaper habit." What set this hypothesis apart from casual observations of newspaper consumption practices was the claim that habit transcended demographic factors as an impetus of newspaper readership. And indeed, the data showed that one could not easily describe a "typical" habitual reader as the effect of habit was present despite the influences of age, education and income. It is worth noting that this statement likely could not be made without the use of regression analysis. The technique allowed the researcher to peel away the layers of influences on newspaper readership

193 in a manner quite improbable by other social scientific methods. To isolate, for instance, readers who are male, between 30 and 40 and with incomes above $40,000 by regulating their access to newspapers is not feasible in the competitive world of western mass media. TABLE 19. Review of Hypotheses
No. Hypothesis Supported / not supported

4 5

Newspaper reading habit will positively influence people to read newspapers even when the effects of demographics and other traditional drivers of readership are controlled. The greater the readership of entertainmentbased content, the more a reader will display newspaper habit. The greater the readership of local newsbased content, the more a reader will display newspaper habit. The greater the readership of obituaries, the more a reader will display newspaper habit. The greater the readership of international news, the less a reader will display newspaper habit. The more readers use newspapers to reinforce their community identity, the more they will display newspaper habit. The more readers use newspapers to make their lives easier (utility), the more they will display newspaper habit.

Supported

Supported

Not supported

Supported Supported, marginally

Supported

Not supported

194 An earlier attempt by Stone (1979) to quantify the existence of the newspaper habit was hampered by serious methodological problems. The survey forms were distributed on a college campus without an attempt to randomize the sample. As college students are, by Stone's own account, in a particular age range, have middle to upper class incomes and (by definition) are highly educated, the 1979 study could not account for the effects of age, income and education. Those three factors, as reported earlier, are already well known for their power to influence newspaper reading. In this study, however, the combination of a random sample drawn over a broad geographical area, the variable-screening capability of the statistical methodology and the use of multiple tests allowed the researcher to rephrase Hypothesis 1 with confidence: There is a newspaper habit that affects readership regardless of age, income and education. The greater difficulty in this line of research is defining or measuring the finer nuances of the newspaper habit. This can be seen in the tests for Hypotheses 2-7. Hypothesis 2 seemed the simplest of these to approach: The greater the readership of entertainment-based content, the more a reader will display newspaper habit. Newspaper folklore is replete with anecdotes about entertainmentbased reading. The term "funny papers" as a description for the comics or

195 cartoon section of the newspaper plays directly to the theme of entertainment-based reading. Journalism texts often uses the term "soft news" to describe entertaining stories about non-breaking events and give it second-ranking to traditional "hard news." (Rich 1997) However, they give begrudging recognition to the popularity of such content with readers. (Brooks et al. 1999) A key to deciphering this variable was the word "entertainment," which can be both an activity and a content goal. For instance, Study 1 found that people who read for entertainment have low levels of newspaper habit. On the other hand, people who read for relaxation (and, logically seek entertaining material to read) have high levels of newspaper habit. This was clarified in Study 2, which found that pages labeled "entertainment" were not favored by people with high levels of habit, but personally entertaining pages such as the comics, living section, crossword puzzle, advice columns, arts section, the horoscope and the TV section were favored more highly by the habit-prone than by low-habit readers. The sum of this mixed analysis is support for Hypothesis 2. Hypothesis 3 also seemed to be based in intuitive logic: The greater the readership of local news-based content, the more a reader will display newspaper habit.

196 Local news is often identified as the most popular subject for newspaper readers, as Layton (1999) wrote in his non-academic but extensive comparison of scholarly and administrative reader research. Even early studies such as that conducted by Berelson (1949) noted the overwhelming popularity of everyday "news." This popularity, however, seemed to also overwhelm any effect of habit. Local news was just as popular with habituated readers as those with no newspaper habit. "News" is just popular. Period. There was no direct support for the notion that it had a habit relationship. The frustratingly logical aspect of the uses and dependency model, however, is the implication of a circular effect between habit and the desirability of news: News is simply a popular content item, so people acquire the habit of reading it through sheer repetitive exposure. When the habit becomes strong, the news content may be read not because the consumers particularly like it, but because it has become their ritual to read it anyway. The "chicken and egg" question of which factor precedes which (or at what point is there a change in effect) is rich fodder for future research. Hypothesis 4 grew out of the personal curiosity of the researcher. Subjects in the 1998 preliminary study repeatedly mentioned their love for the obituaries printed in the rural daily paper they read. This also matched the vaudeville-comic joke "I woke up this morning and read the

197 obituaries to see if I was in them." And indeed, people who exhibited higher levels on newspaper habit ranked the popularity of obituaries much higher than did people with lower levels of habit. The obituary columns of a newspaper seem well suited to encourage habituation. They are run in every edition of the paper and usually on the same page or in the same general location of the paper. They provide both surveillance of the community and voyeuristic entertainment in the form of a close look at the personal lives of community members, yet provide tangible utility in the form of schedules for one of the most popular ritualistic rites of western society the funeral (Carter 1999; Coppet 1992; Imber-Black and Roberts 1992). The support for the hypothesis,

then, has intuitive as well as empirical logic. The logic of Hypothesis 5, however, was less clear a condition that was reflected in the findings: The greater the readership of international news, the less a reader will display newspaper habit. The assumption of the researcher was that international news was the counter to the ever-popular local news and was, therefore, less popular with habitual newspaper readers. Indeed, it ranked slightly higher in

popularity among people with strong newspaper habits than among people with no habits. The hypothesis was supported, but only marginally so.

198 Hypothesis 5 may have succumbed to the same fate as Hypothesis 2: the simply overwhelming popularity of "news." The researcher readily concedes this point and did not launch this study to prove that people read newspapers so they can read news. Rather, the intent was to determine what additional factors encourage readership. From the data, it appears that fine distinctions in the definition of hard news are not among those factors. The last two hypotheses took a different approach and looked at how readers use the news rather than whether they simply enjoyed the news. The two hypotheses provided measures of the dependency relationships that readers form with their newspapers. Hypothesis 6 provided a framework to look at the communitybuilding role of newspapers: The more readers use newspapers to reinforce their community identity, the more they will display newspaper habit. The first test of the hypothesis was a direct question posed in Study 1: "Does reading the newspaper make you feel like part of your community?" People with a high level of newspaper habit were much more likely to answer "yes" than anyone else, supporting the hypothesis. An attempt to add to that support by identifying sections of the paper that amplify community support had only marginal success. The researcher tracked the effect of habit on readers of editorial pages and letters to the

199 editor on the theory that these features provided a forum in which the future of the community was discussed. Readers with high levels of newspaper habit ranked the editorial pages marginally higher, but the letters to the editor marginally lower. The opinionated nature of these pages may contribute to the confusion. Editorial pages and the letters to the editor are among the few overtly subjective portions of a newspaper. In theory, the readers form their own opinions after reading the "objective" or "non-biased" stories in the main part of the paper. It is possible that the overtly subjective nature of the editorial pages causes readers to judge them with a different set of criteria than they do the news pages. Although the data from Study 1 support the hypothesis, future researchers on the topic should identify other newspaper content that contributes to community building. Finally, Hypothesis 7 addresses the issue of utility finding functional value in the content of the newspaper: The more readers use newspapers to make their lives easier (utility), the more they will display newspaper habit. Again, a two-step process was used to test the hypothesis. Study 1 asked a direct question: "Do you read the newspaper to find information to make your life easier?" Then Study 2 sought to verify the results from the use of particular content of the paper.

200 Unfortunately, the answer to the Study 1 question was a simple "yes." Readers of every ilk said they use the newspaper to make their lives easier more so than they used it to build community, provide entertainment or to relax. "Utility" has the same overwhelming characteristics as those of "news." In that overwhelming response, however, readers with low or no habit cited utility more than did those with high habit. However, readers with moderate habit levels cited utility more than readers with low habit but less than people with no habit. A more detailed set of questions and a more sophisticated method of analysis is required to use the direct-response technique to test this hypothesis. The analysis in Study 2 of classified advertising, public records, calendars, weather and other utilitarian information categories soundly rejected Hypothesis 7. It was the readers with low levels of habit who sought utility, not those with high habit. This makes some intuitive sense. Persons with a functional, nonemotional relationship with a newspaper but who still use the paper regularly may view that newspaper as a simple tool. When they need it, it is there as a reference to the data they need to make their lives easier. When they do not need the data, they do not need the newspapers.

201 Habitual readers, however, logically should want the paper whether or not they need it. Habit is more closely associated with emotion, and like emotion supersedes logic. While Hypothesis 7 was not supported, it did point to a media dependency relationship between low-habit readers and the newspaper. For these readers the newspaper truly fills the information gap in urban life filled by family and friends in a more pastoral village life (Merskin 1999). This utility-based dependency relationship should be verified by further research.

Implications

The most immediate implication of this research comes in the form of an answer to the archetypical editor's question in Chapter 1: "Is it really just the news? Is it really just the content I supply that attracts the readers to the newspaper?" The data from this research clearly show that the answer is a conditional "no." Certainly, as indicated in Study 2, the content of a newspaper is extremely important to readers. But some readers seem to regularly peruse their newspapers despite the content, not just because of it. Whether one calls it the newspaper habit or a media ritual, the consumption of newspapers by Americans purely for the sake of enjoyably reading them is a fact of life not to be underestimated.

202 The regression analyses of Studies 1 and 2 indicate that between 6% and 15% of readers' attraction to newspapers the construct of readership can be linked to pure habit unadulterated by the powerful demographic readership drivers of age, education and income. If one rounds this to a marginal figure of 10%, it at first seems like a minor factor in the use of newspapers by the American public. However, Study 2 showed that 55% of the readers had some level of newspaper habit (Table 13). When one combines that figure with the tremendous challenges of declining circulation facing newspaper companies in the United States, a minor figure comes to represent a major opportunity. As reported in Chapter 1, the circulation of American newspapers when seen as a proportion of the country's population has trended downward for many years. Perhaps more disconcerting to newspaper managers is the skyrocketing rate of "churn" as high as 70% at many papers (Giobbe 1994). Churn is an indication of fickle readers they subscribe for a time and then drop their subscription, perhaps only to resubscribe at another date. A churn subscriber might drop the paper and resubscribe several times in a single year to take advantage of new subscriber discounts and bonuses. Habit, on the other hand, is better associated with reader loyalty subscribing to the paper year in and year out.

203 A churn rate of 70% means that for each subscriber a newspaper can count at the beginning of the year, it must find 2.3 "new" subscribers by the end of the year merely to keep even. (For every 100 subscribers, there are 30 loyal readers and 70 churn readers who will drop their subscription at some point. 70/30= 2.3.) Simple math shows the financial impact of this. Although the traditional target for circulation profit is 46% (IPA 1999), let us take the example of a newspaper that prices its subscriptions to produce a modest 20% profit over the cost of distributing the paper. If the marketing costs such as telemarketing crews and subscription bonus "gifts" halve the profit of a new subscription, then each long-term subscriber who does not require expensive marketing each year is "worth" twice the value of a churndriven subscriber. The payoff for instilling readers with a solid newspaper habit is substantial. With our hypothetical 70% churn rate, a paper would have to increase circulation by approximately 10% to earn the same profit it would if it merely dropped churn to 60%. Decreasing churn rather than increasing circulation also returns that increased profit at a substantially higher rate, as there is no additional cost in newsprint, production and distribution as there would be with a 10% increase in the number of papers printed.

204 In reality, the return is probably considerably higher than proposed in this hypothetical scenario even though the churn rate of most newspapers is below the 70% level. The researcher's lengthy career in the newspaper industry taught him that "promotional" subscriptions designed to attract new subscribers are very often sold at no profit or even at a loss. That observation was verified by German media marketing analyst Elke Low (1995), who in a newspaper marketing journal detailed a lengthy menu of costs associated with new or promotional subscriptions. But Low also cited a Swedish study of the public relations benefit of satisfying and retaining customers: Satisfied customers told an average of three friends about the paper. Displeased customers complained to an average of 11 friends. If, as this dissertation demonstrates, the newspaper habit is very real and can help generate profit and stability for newspapers, then it has implications for the internal culture of newspaper companies (E&P 1999; Fitzgerald 1999, 1999). A look at newspaper trade magazines reveals many articles about how to attract "new" readers (E&P 1999; Fitzgerald 1999, 1999). Circulation departments, editorial departments and

advertising departments are like much of the rest of America enthralled by the "new and improved" consumer philosophy. Premiums and prizes are offered to readers for starting a subscription, but seldom for maintaining a subscription. In fact, the promotional subsidies mentioned

205 above make new subscriptions less expensive to the readers than are old or continuing subscriptions. As a media manager, it would be easy to employ the logic apparently followed by many newspapers today concentrate on attracting new subscribers to replace those the papers lose through churn. After all, if readers are loyal and habitual, by definition that should mean that they will continue to read with no care or grooming on the part of the publisher. But as the examples above demonstrate, efforts to increase the proportion of habitual readers to churn-driven readers could offer substantial financial rewards to publishers. In addition, today's popular culture offers myriad opportunities for those readers to take their media habits elsewhere. We live in a mediarich world in which promoters of television, internet, radio, magazines, recordings, video games, movies, books and even lecturers spend millions of dollars to woo consumers. The loyal newspaper readers may be habitual, but they are not guaranteed in perpetuity. However, a commitment to cater to habituated newspaper readers for the sake of profitable subscriber retention would require revisions in the strategic and tactical thinking of newspaper executives. Change in design, content or service could be undertaken only with great care, and "new" would not automatically correlate with "better." The value of an "improvement" for the sake of attracting new readers would hang in the

206 balance against the possible irritation it could cause loyal readers. Obituaries, which many newspapers relegate to second-class status (Brooks et al. 1999), would require the attention of more experienced journalists. In addition, the role of entertaining readers might necessarily be elevated to a level of importance close to that now given to informing readers. This in turn would require a difficult re-examination of the industry's attitude toward "serious news" and "happy talk" so entertainment values are raised without depreciating the already-popular role of news. A final implication concerns the impact of this study on audience theory. The empirical confirmation of newspaper habit leaves little question that an individual's psychological or sociological needs alone are not enough to explain why they read newspapers and, by association, why they consume mass media. Habit is poorly explained if explained at all by the uses and gratifications model or by the media systems dependency model. But if the newspaper habit exists, as this research demonstrates,

then it must be accounted for in the explanation of media use. Convergence of the two theories is one logical answer and is attempted through the uses and dependency model (Rubin and Windahl 1986). However, the model has been subjected to too little scholarly inquiry to elevate it to "theory" status. Either of the two major theories could also be modified to address habit. Full development of uses and dependency or an explanation of newspaper habit through the uses and

207 gratifications model or media systems dependency will take considerable research and theory examination. This dissertation project, though lengthy, is only a beginning. As mentioned previously, good research often produces as many questions for later researchers as it does answers for the primary researcher.

Further Research

Below is a list of recommendations for further research based on this study, followed by a list of recommendations to newspaper publishers based on the data in this dissertation: 1. With the existence of a newspaper habit established, more research is necessary to determine how that habit develops. Is it certain content such as local news that leads to the habit, or is it certain behavior such as sitting in a recliner that leads to the habit? Advice columnists and other "experts" in the popular press commonly state that a behavior repeated for 21 days becomes a habit (Ewer 1999). Can media habits be established in such a short time, giving new value to trial subscriptions and free samples? By the same token, can competitors, governments or others "short circuit" media habits? 2. Just how many habitual readers are out there? The data from this study indicate habit is beteween 6% and 15% of the explanation of why people read newspapers and that 55% of the readers had an average to

208 very strong newspaper habit. Those figures must be fine-tuned and verified to be of substantial use to researchers and the newspaper industry. 3. Do habitual readers share psychological, sociological and psychographic traits? Is the newspaper habit a reaction to other stimuli, a learned behavior or a physiological reaction? 4. Media systems dependency draws a parallel between the media of an urban society and the friends or family in a village society. Uses and dependency theory says personal needs and behaviors such as habits are mixed with the societal dependencies. Could there be a connection between the "village" and the habits? Are media habits, like values, passed on from parent to child or are they learned from the mass media/surrogate parent in an urbanized society? Is there a relationship between habit and its cultural cousin custom? 5. The data seem to support the contention that some types of content contribute to a newspaper habit. Determining the factors within that content that stimulate the habit should be of great interest to both sociologists and journalists. A variation on the theme would explore the impact of the presentation of that content, both graphically and in mode of delivery. 6. Are media habits permanent, or as transitory as skirt lengths and necktie widths? It was relatively easy to identify readers with a

209 newspaper habit, but this research shed little light on when those readers developed the habit and whether they will have it tomorrow. 7. Newspapers around the world are attempting to move their brand identity to the Internet by providing versions of their print editions on the World Wide Web. This effort offers an opportunity to test the power of habit as a medium develops. Can the habits of the newspaper readers be transferred to a new medium with the same ease that ink-on-paper words are transferred to a computer screen? Will new habits of on-line readers affect their on-paper reading?

Recommendations to Newspaper Publishers

Beyond its theoretical base, this research can have practical applications for the newspaper industry. Based on the data, the researcher recommends: 1. Identify your habitual readers. The type of research conducted for this dissertation could not quantify the actual number of habitual readers, but it indicates that habit makes up more than 6% to 15% of the reason people read. The people with strong newspaper habits should be fairly easy to identify and small enough to accommodate. 2. Revel in your habitual readers. Once you have identified them, give them the respect and attention they deserve. Form a "loyal readers

210 club" that offers discounts and premiums, and use your internal resources to promote long-time readers. 3. Grow more habitual readers. If internal research shows many of your habitual readers enjoy the paper while sitting in a recliner, consider giving away promotional subscriptions with sales of new recliners. 4. Protect your habitual readers. Any business can become ethnocentric to the detriment of its customers. The hip comic-page changes passionately advocated by a 25-year-old, unmarried copy editor may not be appreciated by the 55-year-old father of four who reads his "Rex Morgan, M.D." every morning over coffee.

Conclusion

Back in a time when radio was a startlingly new technology and the Internet was not yet even science fiction, the philosopher William James applied the then-familiar vocabulary of mechanics to the concept of habit (James 1914). He said habit was society's "great fly wheel." Just like one

of those heavy, finely balanced wheels, habit seems propelled by inertia alone, spinning along endlessly without attention from the engineer. The analogy is still a good one both for habit itself and for research into habit's role in the mass media. Habit cares not a whit whether it is promoted, researched, coddled or prodded. Its power is that it continues

211 despite the best effort of marketers, despite the proclamations of academics, despite the jokes of late-night TV comedians. Researchers and newspaper managers have largely forgotten habit in their zeal to document the effects of the media and the myriad uses people make of the media. But the fly wheel spins on, producing those puzzling atypical results in surveys and the occasional angry call to newspaper editors. The forgotten part of James' metaphor is that fly wheels not only store kinetic energy, they can supply it. Reports of how people partake of their "newspaper habit" are more than quaint stories for after-dinner speakers. They are instead reports of the way media use is truly shaping the daily lives of people and of how people transform their day-to-day activities into that odd beast we call culture. Observing the wheel spin, mass media researchers and newspaper professionals are faced with the same question: Do we watch, ride or just get out of the way?

212

APPENDIX A

STUDY 1 QUESTIONS

Words and phrases in all capitals are variable names and interviewer instructions which are not read aloud. All questions are allowed dont know, refuse and no answer categories, but these are excluded to save space. Words and phrases in parentheses, as well as probes, are used at the interviewers discretion. This survey has several questions that have two versions of answer categories: one version is randomly assigned to half all respondents and the second version is randomly assigned to the other half. These are all designated with a note in CAPS before them. The following questions are asked of all 800 respondents: the ORPARK series, SMOKE series, GAMBLE/CASINO series, and the demographic questions. Seven other groups of questions are asked randomly of 400 respondents each. The first 400 respondents are asked the SCH series, NEWS series, and ECON series. After ECON6, the first 400 respondents skip to SEX. Note: those who skip out of the ECON series at ECON1, also skip to SEX. The second 400 respondents are asked the MEAS64 series, FOOD series, the single question PURCH, the ENVIR series, and the RELIG series. The skip logic here gets a little confusing. For the second 400 respondents, those who dont gamble (or who dont gamble in Oregon, or who volunteer that they are casino employees), ae skipped to MEAS64A from GAMBLE1, GAMBLE2 or GAMBLE3. Persons who go through the entire GAMBLE/CASINO series also will need to skip to MEAS64A.

HELLO1 Hello. This is ____ calling from the University of Oregon Survey Research Laboratory. We are conducting a 10-minute survey of adults age 18 or older for the Oregon Parks Department, the University of Oregon, and two private research organizations (Economics Consulting Northwest in Portland and Decision Research in Eugene) about your opinions on a

213 variety of other issues. I want to assure you that I am not selling a thing, and that this survey is completely anonymous and voluntary. Please do not even tell me your name. 1 R ON TELEPHONE CTRL/END --> SCHEDULE CALLBACK HELLO2 Do you have any questions about the survey before we begin? 1 NO QUESTIONS OR QUESTIONS ANSWERED, OK TO BEGIN SURVEY HAS QUESTIONS ---> REFER TO INTERVIEWER INSTRUCTIONS COOPERAT We appreciate your cooperation. (I'd like to begin the survey now.) 1 OK CTRL/END NO ORPARK1 I will begin by asking you some questions about Oregon parks. Is it important to you for Oregon to have public campgrounds? 1 YES 2 NO ORPARK2 Is it important to you for Oregon to provide public beaches that are easily accessible? PROBE: easy to get to? 1 YES 2 NO ORPARK3 Is it important to you for Oregon to have well-maintained trails for hiking, biking, and horseback riding? 1 YES 2 NO PROGRAMMER: IF NONE WAS MARKED IMPORTANT IN ORPARK1, ORPARK2 AND ORPARK3, SKIPTO ORPARK5 IF ONE WAS MARKED IMPORTANT IN ORPARK1, ORPARK2 AND ORPARK3, AUTOMATICALLY CODE IT IN ORPARK4 AND SKIPTO ORPARK5 IF TWO OR MORE WERE MARKED IMPORTANT IN ORPARK1, ORPARK2 AND ORPARK3, HIGHLIGHT THEM IN ORPARK4.

214

ORPARK4 Which one of these is most important to you? PROBE FROM HIGHLIGHTED ITEMS ON LIST 1 PUBLIC CAMPGROUNDS 2 ACCESSIBLE PUBLIC BEACHES 3 WELL-MAINTAINED TRAILS ORPARK5 Is it important to you for Oregon to provide family-oriented outdoor activities, such as nature trails, evening programs at campground, and guided nature walks? 1 YES 2 NO ORPARK6 Is it important to you to preserve Oregons historical places? PROBE: Places such as Fort Stevens, historic districts (in some cities), and historic trails (like the Oregon Trail, (the Applegate Trail, and the Santiam Trail)). 1 YES 2 NO ORPARK7 Is it important to you that Oregonians learn about the states forests, beaches, waterways, and wildlife? 1 YES 2 NO PROGRAMMER: IF NONE WAS MARKED IMPORTANT IN ORPARK5, ORPARK6 AND ORPARK7, SKIPTO ORPARK5 IF ONE WAS MARKED IMPORTANT IN ORPARK5, ORPARK6 AND ORPARK7, AUTOMATICALLY CODE IT IN ORPARK4 AND SKIPTO ORPARK5 IF TWO OR MORE WERE MARKED IMPORTANT IN ORPARK5, ORPARK6 AND ORPARK7, HIGHLIGHT THEM IN ORPARK4. ORPARK8 Which one of these is most important to you? PROBE FROM HIGHLIGHTED ITEMS ON LIST 1 FAMILY ORIENTED OUTDOOR ACTIVITIES 2 HISTORICAL PLACES STAYING PRESERVED

215 3 LEARNING ABOUT NATURE ORPARK9 Do you think that building next to Oregons rivers should be restricted, within certain guidelines? 1 YES 2 NO ORPARK10 Overall, would you rate Oregons state parks as excellent, good, fair or poor? 1 EXCELLENT 2 GOOD 3 FAIR 4 POOR 5 IF VOLUNTEERED: IT VARIES ORPARK11 How often did you go camping when you were a child -- often, sometimes, or never? PROBE: by child, I mean less than 18 years old. 1 OFTEN 2 SOMETIMES 2 NEVER ORPARK12 Have you been camping in the past 12 months (since November 1997)? 1 YES 2 NO --> SKIPTO SMOKE1 3 IF VOLUNTEERED: NEVER CAMPED IN LIFE --> SKIPTO SMOKE1 ORPARK13 Have you been camping *in Oregon* in the past 12 months (since November 1997)? 1 YES 2 NO ORPARK14 When you go camping, do you camp mainly in public campgrounds or privately-owned campgrounds? PROBE: Public campgrounds are owned and operated by citizens. (For example, state campgrounds are owned and operated by the state of Oregon.)

216 PROBE: Private campgrounds are owned and operated by private people or groups of people. Private campgrounds try to earn a profit; public campgrounds are nonprofit. 1 PUBLIC 2 PRIVATE 3 IF VOLUNTEERED: BOTH ORPARK15 Do you camp mainly in a tent, a recreational vehicle, or something else? 1 TENT 2 RV 3 SOMETHING ELSE ORPARK16 Compared to five years ago, are you camping more often, less often, or about the same? 1 MORE 2 LESS 3 ABOUT THE SAME ORPARK17 Do you think public campground fees in Oregon are too high, too low, or just about right? 1 TOO HIGH 2 TOO LOW 3 JUST ABOUT RIGHT ORPARK18 Do you think Oregons public campgrounds are usually crowded or uncrowded? PROBE FOR IT DEPENDS: Overall... 1 CROWDED 2 UNCROWDED 3 IF VOLUNTEERED: IT VARIES ORPARK19 Do you think (they/Oregons public campgrounds) are usually kept up or run-down? PROBE FOR IT DEPENDS: Overall... 1 KEPT UP 2 RUNDOWN 3 IF VOLUNTEERED: IT VARIES

217 ORPARK20 Do you prefer public campgrounds that look like a city park, or do you prefer a natural look? PROBE: Do you prefer a landscaped look or a natural look in public campgrounds? PROBE FOR IT DEPENDS: Overall... 1 LANDSCAPED 2 NATURAL 3 IF VOLUNTEERED: IT VARIES SMOKE1 The next few questions are about smoking and about visiting casinos. Have you ever regularly smoked or used tobacco? PROBE: By regularly, I mean using tobacco at least once a day for 30 days. NOTE: INCLUDE SMOKELESS TOBACCO, CHEW, SNUFF, PIPES, CIGARS 1 YES 2 NO --> SKIPTO GAMBLE1 SMOKE2 Do you smoke or use tobacco regularly now? PROBE: By regularly, I mean using tobacco at least once a day for 30 days. PROBE: Do you smoke cigarettes or do you use some other kind of tobacco (or both)? NOTE: INCLUDE SMOKELESS TOBACCO, CHEW, SNUFF, PIPES, CIGARS 1 YES, CIGARETTES 2 YES, OTHER TOBACCO 3 YES, BOTH CIGARETTES AND OTHER TOBACCO 4 NO --> SKIPTO SMOKEAGN SMOKEFRQ (On average) how many cigarettes do you smoke per day? PROBE: Less than half a pack, between one half and one whole pack a day, or more than one pack a day? 1 LESS THAN HALF A PACK PER DAY (10 OR FEWER CIGARETTES) 2 BETWEEN HALF AND ONE PACK PER DAY (11-20 CIGARETTES) 3 MORE THAN ONE PACK PER DAY (21 OR MORE CIGARETTES) SMOKEAGN If you could go back to the time when you first began to smoke, would you decide to smoke again? 1 YES 2 NO

218 SMOKEAG How old were you when you first started (smoking/using tobacco) regularly? PROBE: By regularly, I mean using tobacco at least once a day for 30 days. TYPE EXACT AGE BELOW, 5-96 GAMBLE1 Have you played bingo or gambled at a casino in the past 12 months? PROBE: since November 1997? 1 YES 2 NO --> SKIPTO SCH1 3 IF VOLUNTEERED: R IS A CASINO EMPLOYEE --> SKIPTO SCH1 GAMBLE2 Have you played bingo or gambled at a casino *in Oregon* in the past 12 months? PROBE: since November 1997? 1 YES 2 NO --> SKIPTO GAMBLE3 3 IF VOLUNTEERED: R IS A CASINO EMPLOYEE --> SKIPTO SCH1 GAMBLE2A How many times (did you visit Oregon casinos and play Bingo or gamble) (in the past 12 months)? NOTE: R MAY HAVE MADE MULTIPLE VISITS AT MULTIPLE CASINOS IN ONE TRIP. NOTE: INCLUDE UNPLANNED DROP-IN, PASSERBY VISITS. PROBE: since November 1997? PROBE: How many different times have you gone to Oregon casinos to play Bingo or gamble? CODE EXACT NUMBER 1-996 GAMBLE2B How much money did you usually set aside for gambling in Oregon each trip? PROBE: How much money did you plan to spend, even if it is different from how much you actually spent? PROBE FOR DROP-INS: When you dropped in, how much money did you set aside? NOTE: THE MONEY SET ASIDE MAY BE DIFFERENT FROM MONEY ACTUALLY SPENT. CODE EXACT DOLLARS, 0-99995

219 99996 R DID NOT SET ASIDE AN AMOUNT OF MONEY CASINO1 I am going to name some casinos in Oregon. Please tell me if you have played at each one in the past 12 months. The first one is Spirit Mountain. PROBE: The Spirit Mountain Casino is on Highway 18, between McMinnville and the Oregon coast. 1 YES 2 NO CASINO2 (The next one is ...) Wildhorse. PROBE: The Wildhorse Casino is off of Interstate 84 in Pendleton. 1 YES 2 NO CASINO3 (The next one is ...) Chinook Winds. PROBE: The Chinook Winds Casino is on Highway 101 in Lincoln City (on the Oregon coast). 1 YES 2 NO CASINO4 (The next one is ...) Cow Creek or Seven Feathers. PROBE: The Cow Creek and Seven Feathers Casinos are on Interstate 5 (between Roseburg and Medford). 1 YES 2 NO CASINO5 Did you play at any other casinos in Oregon in the past 12 months (since November 1997)? 1 YES 2 NO --> SKIPTO CASINO10 CASINO6 (Did you play at) the Mill Casino? PROBE: The Mill Casino is near Coos Bay (on the Oregon coast). 1 YES 2 NO

220 CASINO7 (Did you play at the) Indian Head (Casino)? PROBE: The Indian Head Casino is at the Kah-Nee-Ta Resort (in central Oregon, north of Bend). 1 YES 2 NO CASINO8 (Did you play at the) Kla-Mo-Ya (Casino)? PROBE: The Kla-Mo-Ya Casino is on Highway 97 (between Bend and Klamath Falls). 1 YES 2 NO CASINO9 (Did you play at the) Old Camp (Casino)? PROBE: The Old Camp Casino is near Burns (in eastern Oregon). 1 YES 2 NO CASINO10 What do you like best about Oregon casinos? PROBE: Think about the Oregon casino you have been to the most. (What do you like best about (it/that one)? PROBE: Why? PROBE: Is there anything else? OPEN-ENDED CASINO11 If there is one thing you could change or improve about Oregon casinos, what would that be? PROBE: Think about the Oregon casino you have been to the most. (What would you change or improve in (it/that one)? PROBE: Why? PROBE: Is there anything else? OPEN-ENDED GAMBLE3 Did you play Bingo or gamble at a casino outside of Oregon in the past 12 months? 1 YES 2 NO --> SKIPTO SCH1

221 GAMBLE3A How many trips did you make to play at casinos outside of Oregon in the past 12 months? PROBE: since November 1997? PROBE: How many times have you gone to casinos outside of Oregon (to play Bingo or gamble)? NOTE: INCLUDE UNPLANNED, DROP-IN, PASSERBY VISITS. NOTE: HERE WE WANT TRIPS, NOT CASINO VISITS. INCLUDE TRIPS WHERE CASINOS AND GAMBLING WERE SECONDARY ACTIVITIES. CODE EXACT NUMBER 1-996 GAMBLE3B How much money did you usually set aside for each trip gambling outside of Oregon? PROBE: How much money did you plan to spend, even if it is different from how much you actually spent? PROBE FOR DROP-INS: When you dropped in, how much money did you set aside? NOTE: THE MONEY SET ASIDE MAY BE DIFFERENT FROM MONEY ACTUALLY SPENT. CODE EXACT DOLLARS, 0-99996 99996 R DID NOT SET ASIDE AN AMOUNT OF MONEY SCH1 The next few questions are about schools. Do you have any children age 18 or younger in your home? 1 YES 2 NO --> SKIPTO SCH3A OR SCH3B SCH2 Do you have a child in public school? 1 YES 2 NO (RANDOMLY ASSIGN HALF OF ALL RS TO SCH3A AND HALF TO SCH3B) SCH3A Currently, would you rate Oregons public schools as excellent, good, fair or poor? 1 EXCELLENT 2 GOOD

222 3 FAIR 4 POOR 5 IF VOLUNTEERED: IT VARIES SCH3B Currently, would you rate Oregons public schools as poor, fair, good, or excellent? 1 POOR 2 FAIR 3 GOOD 4 EXCELLENT 5 IF VOLUNTEERED: IT VARIES (RANDOMLY ASSIGN HALF OF RS TO SCH4A AND HALF TO SCH4B) SCH4A Do you think the standards for the quality of Oregon public schools should be set at the national, state or local level? 1 NATIONAL 2 STATE 3 LOCAL 4 IF VOLUNTEERED: OTHER SCH4B Do you think the standards for the quality of Oregon public schools should be set at the local, state or national level? 1 LOCAL 2 STATE 3 NATIONAL 4 IF VOLUNTEERED: OTHER SCH5 Overall, in your opinion, are Oregons public schools doing a good job preparing children for adult roles? 1 YES 2 NO SCH6 Have you heard about the School Reform Act which creates new standards for Oregon school children? 1 YES 2 NO --> SKIPTO NEWS1

223 SCH7 Do you think (this/ the School Reform Act) will have mostly good or mostly bad results for Oregon? 1 MOSTLY GOOD 2 MOSTLY BAD --> SKIPTO SCH7B SCH7A Why (do you think it will have mostly good results)? OPEN ENDED --> SKIPTO NEWS1 SCH7B Why (do you think it will have mostly bad results)? OPEN ENDED NEWS1 Now, thinking about the news. How important is it to you to read a newspaper regularly -- very important, somewhat important, or not important? PROBE: Is it very important, somewhat important, or not important? 1 VERY IMPORTANT 2 SOMEWHAT IMPORTANT 3 NOT IMPORTANT --> SKIPTO ECON1 NEWS2 Does reading the newspaper make you feel like part of your community? 1 YES 2 NO NEWS3 Do you read the newspaper for entertainment? 1 YES 2 NO NEWS4 Do you read the newspaper for relaxation? 1 YES 2 NO NEWS5 Do you read the newspaper to find information to make your life easier? 1 YES 2 NO

224

IF NEWS2 EQ 2, NEWS3 EQ 2, NEWS4 EQ 2, AND NEWS5 EQ 2, SKIPTO NEWS7 ELSE HIGHLIGHT THE ANSWERS BELOW THAT R ANSWERED YES. NEWS6 Which one of these reasons for reading a newspaper is most important to you? PROBE FROM HIGHLIGHTED ITEMS ON LIST 1 FEELING PART OF COMMUNITY 2 ENTERTAINMENT 3 RELAXATION 4 FINDING INFORMATION TO MAKE LIFE EASIER NEWS7 Do you usually read the newspaper at the same time of the day (when you read it)? 1 YES 2 NO NEWS8 Do you usually read the newspaper in the same place (when you read it)? PROBE: Such as, the same chair, at the same table, on the bus, at your desk, or in the car? 1 YES 2 NO NEWS9 Do you usually have the same thing to drink or eat while you are reading the newspaper? 1 YES 2 NO ECON1 Do you pay attention to the international economy in the news? 1 YES 2 NO --> SKIPTO SEX ECON2 Do you approve or disapprove of how the President is handling the overall economy? 1 APPROVE 2 DISAPPROVE

225

ECON3 Do you approve or disapprove of how the President is handling international trade? 1 APPROVE 2 DISAPPROVE (RANDOMLY ASSIGN HALF OF ALL RS TO ECON4A AND HALF TO ECON4B) ECON4A Overall, do you think international trade is good for the American economy, bad for the American economy, or has no effect? 1 GOOD 2 BAD 3 NO EFFECT ECON4B Overall, do you think international trade is bad for the American economy, good for the American economy, or has no effect? 1 BAD 2 GOOD 3 NO EFFECT (RANDOMLY ASSIGN HALF OF ALL RS TO ECON5A AND HALF TO ECON5B] ECON5A Overall, do you think international trade is good for Oregons economy, bad for Oregons economy, or has no effect? 1 GOOD 2 BAD 3 NO EFFECT ECON5B Overall, do you think international trade is bad for Oregons economy, good for Oregons economy, or has no effect? 1 BAD 2 GOOD 3 NO EFFECT (RANDOMLY ASSIGN HALF OF ALL RS TO ECON6A AND HALF TO ECON6B)

226

ECON6A Do you think the recent Asian economic crisis has benefited Oregon, has hurt Oregon, or has had no effect? 1 BENEFITED 2 HURT 3 NO EFFECT ECON6B Do you think the recent Asian economic crisis has had no effect on Oregon, has hurt Oregon, or has benefited Oregon? 1 NO EFFECT 2 HURT 3 BENEFITED MEAS64A Did you hear about Measure 64 on the ballot November 3rd, which sought to limit clear cutting and stop chemical sprays in public forests? 1 YES 2 NO --> SKIPTO FOOD1 MEAS64B From what source did you learn most about Measure 64? PROBE FROM LIST 1 BROADCAST MEDIA: TV, RADIO 2 PRINT MEDIA: NEWSPAPER, MAGAZINE 3 IN PERSON: CONVERSATIONS WITH FRIENDS, FAMILY, WORK, CIVIC ORGANIZATION/GROUP 4 DIRECT MAIL, VOTER HANDBOOK 5 BILLBOARD, SIGN 6 OTHER MEAS64C Did you vote for or against this measure? 1 FOR 2 AGAINST 3 IF VOLUNTEERED: DID NOT VOTE MEAS64D Why did you vote against (it/Measure 64)? OPEN-ENDED FOOD1

227 Now a few questions about food. When you buy canned or packaged foods, how regularly do you read the ingredients or nutrition label -- often, sometimes, or never? PROBE: often, sometimes, or never? 1 OFTEN 2 SOMETIMES 3 NEVER FOOD2 How regularly do you buy organic foods -- often, sometimes, or never? PROBE: Organic foods have no chemicals added and are raised without chemical fertilizers. PROBE: often, sometimes, or never? 1 OFTEN --> SKIPTO FOOD3 2 SOMETIMES --> SKIPTO FOOD3 3 NEVER FOOD2A Why dont you buy organic food? OPEN-ENDED FOOD3 Do you eat meat often, sometimes, or never? PROBE: Meat includes chicken, fish, turkey, pork, beef and game meats (from hunting). 1 OFTEN 2 SOMETIMES 3 NEVER FOOD4 Are you concerned with the way meat is raised, butchered, handled, or packaged? 1 YES 2 NO ---> SKIPTO PURCH FOOD4A What are your main concerns? OPEN-ENDED PURCH When you buy things at the store, do you usually think of the impact the things you buy have on the environment? PROBE: When you buy anything from bread to blue jeans to bicycles.

228 1 YES 2 NO ENVIR1 People have different views on the environment. Do you think the environment will take care of itself, no matter what we do to it? 1 YES 2 NO ENVIR2 Do you think the environment needs to be managed by people in order to stay healthy? PROBE: Whatever managed means to you. 1 YES 2 NO ENVIR3 Do you think it is possible for pollution levels to get so high that the environment cannot recover? 1 YES 2 NO ENVIR4 What do you think is more important -- economic growth, even if it leads to environmental problems, or protecting the environment, even if it costs jobs? 1 ECONOMIC GROWTH 2 PROTECTING ENVIRONMENT 3 IF VOLUNTEERED: A BALANCE THE TWO 4 IF VOLUNTEERED: IT DEPENDS ENVIR5 Which of the following statements best fits your views: most corporations place profits ahead of environmental concerns, most corporations place environmental concerns ahead of profit, or most corporations balance profits and environmental concerns? 1 PROFIT AHEAD OF ENVIRONMENTAL CONCERNS 2 ENVIRONMENTAL CONCERNS AHEAD OF PROFIT 3 BALANCE THE TWO ENVIR6

229 How important is protecting the environment to you, personally -- very important, somewhat important, not very important, or not at all important? 1 VERY IMPORTANT 2 SOMEWHAT IMPORTANT 3 NOT VERY IMPORTANT 4 NOT AT ALL IMPORTANT ENVIR7 Who do you think should bear the burden of repairing the environment? OPEN-ENDED RELIG1 The next few questions are about religion. Are you actively involved in a religious or spiritual organization in your community? PROBE: (Id like to remind you,) this survey is completely anonymous. 1 YES 2 NO RELIG2 Are you personally religious or spiritual? PROBE: (Id like to remind you,) this survey is completely anonymous. 1 YES 2 NO --> SKIPTO SEX RELIG3 Does your religion or spiritual beliefs ask you to actively transform the world? 1 YES 2 NO --> SKIPTO RELIG4 RELIG3A How (does your religion or spiritual beliefs ask you to actively transform the world)? OPEN-ENDED RELIG4 Is it important to you to take part in religious or spiritual activities, either alone or in groups? 1 YES 2 NO SEX

230 The last few questions are about yourself. (This may sound silly but) are you male or female? 1 MALE 2 FEMALE AGE How old are you? OPEN-ENDED, ENTER EXACT AGE 18-96 RACE What is your race? PROBE FROM LIST: Are you ... 1 WHITE/CAUCASIAN 2 BLACK/AFRICAN AMERICAN 3 ASIAN AMERICAN/PACIFIC ISLANDER 4 LATINO, HISPANIC 5 AMERICAN INDIAN/NATIVE AMERICAN 6 ESKIMO, ALEUT, ALASKAN NATIVE 7 IF VOLUNTEERED: MIXED RACE 8 IF VOLUNTEERED: OTHER EDUC What is the highest level of education you have completed? PROBE FROM LIST 1 LESS THAN A HIGH SCHOOL DIPLOMA OR GED 2 HIGH SCHOOL DIPLOMA OR GED 3 SOME COLLEGE, NO DEGREE 4 ASSOCIATES DEGREE (AA, AS) 5 BACHELORS DEGREE (BA, BS, AB) 6 MASTERS DEGREE (MA, MS, MBA) 7 DOCTORAL DEGREE (PHD, JD, MD, DDS) 8 OTHER URB_RUR Which of the following best describes the area in which you live? PROBE FROM LIST: 1 A LARGE CITY 2 THE SUBURB OF A LARGE CITY 3 A MEDIUM SIZED CITY 4 THE SUBURB OF A MEDIUM SIZED CITY 5 A SMALL CITY OR SUBURB 6 A SMALL TOWN 7 A RURAL AREA

231

NEWSPAP Do you subscribe a daily newspaper, a weekly newspaper, or both? PROBE: Is that a daily newspaper, a weekly newspaper, or both? 1 YES, DAILY 2 YES, WEEKLY 3 YES, BOTH DAILY AND WEEKLY 4 NO LIB_CONS1 On economic issues, would you describe yourself as liberal, conservative, moderate or something else? 1 LIBERAL 2 CONSERVATIVE 3 MODERATE 4 IF VOLUNTEERED: RADICAL 5 IF VOLUNTEERED: OTHER LIB_CONS2 On social issues, would you describe yourself as liberal, conservative, moderate or something else? 1 LIBERAL 2 CONSERVATIVE 3 MODERATE 4 IF VOLUNTEERED: RADICAL 5 IF VOLUNTEERED: OTHER INCOME4 Is your total annual household income, from all sources, before taxes, over or under $35,000? PROBE: Include money from jobs (wages, salary, tips, bonuses), interest, dividends, child support, alimony, welfare, social security, disability and retirement payments, net income from a business, farm or rent, or any other money income received by members of your family. Do not include lump-sum payments, such as money from an inheritance or sale of a home. 1 OVER $35,000 ---> SKIPTO INCOME3 2 UNDER $35,000 INCOME2 Is it over or under $25,000? PROBE: Is your total annual household income, from all sources, before taxes, over or under $25,000?

232 1 OVER $25,000 ---> SKIPTO ZIPCODE 2 UNDER $25,000 INCOME1 Is it over or under $15,000? 1 OVER $15,000---> SKIPTO ZIPCODE 2 UNDER $15,000---> SKIPTO ZIPCODE INCOME3 Is it over or under $50,000? 1 OVER $50,000 2 UNDER $50,000---> SKIPTO ZIPCODE INCOME5 Is it over or under $75,000 1 OVER $75,000 2 UNDER $75,000 INCOME RECODE IF INCOME1 EQ 2, INCOME EQ 1 (Under $15,000) IF INCOME1 EQ 1, INCOME EQ 2 $15,000 to $24,999 IF INCOME2 EQ 2, INCOME EQ 3 $25,000 to $34,999 IF INCOME3 EQ 2, INCOME EQ 4 $35,000 to $49,999 IF INCOME5 EQ 2, INCOME EQ 5 $50,000 to $74,999 IF INCOME5 EQ 1, INCOME EQ 6 $75,000 or over ZIPCODE What is your zip code? OPEN-ENDED, ENTER EXACT FIVE DIGITS ENDING That is the end of the survey. On behalf of the sponsors of this survey, we thank you sincerely for your time and opinions on these questions. NOQAL Im sorry - ??????????????????????. Good-bye.

233

APPENDIX B

STUDY 2 QUESTIONS

Q:HELLO1 T: Hello, this is ______calling from the University of Oregon Survey Research Laboratory. We are conducting a 15-minute survey of Oregon adults age 18 or older for the Oregon State Library and several researchers at the University of Oregon and Oregon State University about your opinion on a wide variety of issues. I want to assure you that I am not selling a thing, and that this survey is completely anonymous and voluntary. Please do not even tell me your name. Do you have any questions before we begin? Hello, May I please speak with _____________? I'm ________ calling back from the University of Oregon Survey Research Lab to finish an interview we began earlier. Q:PAPER1 T: The next set of questions are about newspapers. In the average week how often do you normally read a newspaper? 1 EVERY DAY 2 4-6 DAYS A WEEK 3 1-3 DAYS A WEEK 4 SELDOM 5 NEVER --> SKIPTO READ1 7 REFUSED 8 DON'T KNOW 9 NO ANSWER IF (ANS > 4) SKIPTO READ1 Q:PAPER2 T: (When you read the newspaper) Do you usually read it at the same time of the day?

234 1 2 7 8 9 YES NO REFUSED DON'T KNOW NO ANSWER

Q:PAPER3 T: (When you read the newspaper) Do you usually read it at the same location? 1 2 7 8 9 YES NO REFUSED DON'T KNOW NO ANSWER

Q:PAPER4 T: (When you read the newspaper) Do you usually read the pages of the newspaper in a particular order? 1 2 7 8 9 YES NO REFUSED DON'T KNOW NO ANSWER

Q:PAPER5 T: I am going to read you seven statements that some people have said describe how they feel when something prevents them from reading the newspaper. Please tell me if any of these apply to you. 1 2 3 4 5 6 7 8 I FEEL INCONVENIENCED I FEEL UNCOMFORTABLE I WORRY THAT I AM "MISSING OUT" ON SOMETHING I FEEL THAT MY ROUTINE HAS BEEN DISRUPTED I FEEL ANGRY I FEEL CONFUSED I REALLY DON'T MISS THE PAPER ON THOSE DAYS EXIT HERE

235 Q:PAPER6 T: When something prevents you from reading the paper as usual do you keep it to read later, tear out certain articles to read later, throw it away or recycle it, give it away, or do something else? 1 KEEP IT 2 TEAR OUT ARTICLES 3 THROW AWAY/RECYCLE 4 GIVE AWAY 5 OTHER (SPECIFY) 7 REFUSED 8 DON'T KNOW Q:PAPER7 T: What page, article or other part of the newspaper is the most important for you to read each time you read the paper? OPEN ENDED Q:PAPER8 T: What part of the newspaper is the second most important for you to read each time you read the paper? OPEN ENDED Q:PAPER9 T: What part of the newspaper is the third most important for you to read each time you read the paper? OPEN ENDED Q:PAPER10 T: Some people consider the newspaper something very important in their life or that they have a "newspaper habit." On a scale of 1 to 5, with 1 being no habit at all, and 5 being a very strong habit, how strong is your need to read a newspaper? 1 NO HABIT 2

236 3 4 5 VERY STRONG HABIT 7 REFUSED 8 DON'T KNOW 9 NO ANSWER Q:SEX T: Thank you for your cooperation so far. To finish the survey I'd like to ask few questions about yourself. Are you male or female 1 2 7 8 9 MALE FEMALE REFUSED DON'T KNOW NO ANSWER

Q:AGE T: In what year were you born? ENTER LAST TWO DIGITS BELOW 96 96 OR MORE 97 REFUSED 98 DON'T KNOW 99 NO ANSWER Q:EDUC T: What is the highest level of education you have completed? PROBE FROM LIST 1 LESS THAN HIGH SCHOOL DIPLOMA 2 HIGH SCHOOL DIPLOMAS/GED 3 SOME COLLEGE, NO DEGREE OR ASSOCIATE'S DEGREE 5 BACHELOR'S DEGREE 6 MASTERS DEGREE 7 DOCTORATE OR PROFESSIONAL DEGREE 97 REFUSED 98 DON'T KNOW 99 NO ANSWER Q:HHNUM

237 T: Including yourself, how many people currently live in your household? PROBE: Include children and people that live there half of the year or more. ENTER EXACT NUMBER 1-9 6 SIX PEOPLE OR MORE 7 REFUSED 8 DON'T KNOW 9 NO ANSWER Q:HHNUM2 T: Do any children under 18 years of age live in your household? 1 YES 2 NO 7 REFUSED 8 DON'T KNOW 9 NO ANSWER Q:MEDICAL T: How would you rate the quality of medical care that is available to you and your family? Excellent, good, fair or poor? 1 EXCELLENT 2 GOOD 3 FAIR 4 POOR 5 NO MEDICAL CARE AVAILABLE (IF VOLUNTEERED) 7 REFUSED 8 DON'T KNOW 9 NO ANSWER Q:URB_RUR T: Which of the following best describes the area in which you live? PROBE FROM LIST: 1 A LARGE CITY OF 250,000 OR MORE 2 THE SUBURB OF A LARGE CITY 3 A MEDIUM SIZED CITY OF 50,000 TO 250,000 4 THE SUBURB OF A MEDIUM SIZED CITY 5 A SMALL CITY OF 10,000 TO 50,000 6 SUBURB OF A SMALL CITY 7 A SMALL TOWN OF 2,500 TO 10,000

238 8 97 98 99 A RURAL AREA (INCLUDES TOWNS OF 2,500 OR LESS) REFUSED DON'T KNOW NO ANSWER

Q:LABFORCE T: What were you doing most of last week -- were you working for pay, keeping house, going to school, looking for work, unable to work, retired, or something else? PROBE FROM LIST 1 WORKING, OR SICK/ON VACATION FROM REGULAR JOB 2 KEEPING HOUSE 3 GOING TO SCHOOL, ON BREAK FROM SCHOOL 4 LOOKING FOR WORK / UNEMPLOYED 5 UNABLE TO WORK, DISABLED 6 RETIRED 8 DOING NOTHING, HANGING OUT AND NOT LOOKING 9 TEMPORARY LAYOFF, LEAVE OF ABSENCE WITHOUT PAY 10 VOLUNTEER WORK ONLY 11 OTHER (SPECIFY) 97 REFUSED 98 DON'T KNOW 99 NO ANSWER Q:OCCIND T: What is your current occupation and industry? PROBE: What do you do or make there? PROBE: What does your company do or make? OPEN-ENDED: TYPE IN EXACT RESPONSE BELOW Q:COMMODE T: How did you usually get to (work/school) last week? 1 Car, truck or van 2 Bus, streetcar 3 Light rail or MAX 4 Taxi 5 Motorcycle 6 Bicycle 7 Walk

239 8 Worked at home or telecommute 9 Equal combination of modes 10 Other 97 REFUSED 98 DON'T KNOW 99 NO ANSWER Q:COMDIST T: How far is your commute? NOTE: ROUND TRIP, IN MILES 96 96 OR MORE 97 REFUSED 98 DON'T KNOW 99 NO ANSWER Q:YEARS T: How many years have you lived in Oregon during your current residency? TYPE IN EXACT NUMBER 96 96 OR MORE 97 REFUSED 98 DON'T KNOW 99 NO ANSWER Q:ZIPCODE T: What is your zip code? OPEN-ENDED, ENTER EXACT FIVE DIGITS TYPE EXACT RESPONSE BELOW Q:RELIGION T: Would you say that religious faith is currently not important, somewhat important, very important or extremely important in your own life? 1 2 3 4 NOT IMPORTANT SOMEWHAT IMPORTANT VERY IMPORTANT EXTREMELY IMPORTANT

240 7 REFUSED 8 DON'T KNOW 9 NO ANSWER Q:VOTER1 T: Are you a registered voter? 1 YES 2 NO ---> SKIP TO INCOME1 7 REFUSED 8 DON'T KNOW 9 NO ANSWER Q:VOTER2 T: Did you vote in all three of the last elections, including the recent vote by mail election? 1 2 7 8 9 YES NO REFUSED DON'T KNOW NO ANSWER

Q:VOTER3 T: Are you a registered Republican, Democrat, Independent, or something else? 1. REPUBLICAN 2. DEMOCRAT 3. INDEPENDENT 4. SOMETHING ELSE (SPECIFY) _____________ 7 REFUSED 8 DON'T KNOW 9 NO ANSWER Q:ENVGRP T: Do you belong to any of the following environmental groups or associations? NOTE: ENTER ALL THAT APPLY PROBE FROM LIST:

241

1 GREENPEACE 2 EARTH FIRST 3 SIERRA CLUB 4 AUDUBON SOCIETY 5 OTHER - SPECIFY, BUT DO NOT PROBE 6 EXIT - (EXIT IF R IS NOT A MEMBER OF ANY OF THE ABOVE) I: Q:INCOME1 T: Is your total annual household income, from all sources, before taxes, over or under $35,000? PROBE: Include money from jobs (wages, salary, tips, bonuses), interest, dividends, child support, alimony, welfare, social security, disability and retirement payments, net income from a business, farm or rent, or any other money income received by members of your family. Do not include lump-sum payments, such as money from an inheritance or sale of a home. 1 OVER $35,000 ---> SKIPTO INCOME4 2 UNDER $35,000 7 REFUSED 8 DON'T KNOW 9 NO ANSWER Q:INCOME2 T: Is it over or under $25,000? PROBE: Is your total annual household income, from all sources, before taxes, over or under $25,000? 1 OVER $25,000 -->SKIPTO ENDIT 2 UNDER $25,000 7 REFUSED 8 DON'T KNOW 9 NO ANSWER Q:INCOME3 T: Is it over or under $15,000? 1 OVER $15,000 ---> SKIPTO ENDIT 2 UNDER $15,000 ---> SKIPTO ENDIT 7 REFUSED 8 DON'T KNOW

242 9 NO ANSWER Q:INCOME4 T: Is it over or under $50,000? 1 OVER $50,000 2 UNDER $50,000---> SKIPTO ENDIT 7 REFUSED 8 DON'T KNOW 9 NO ANSWER Q:INCOME5 T: Is it over or under $75,000 1 OVER $75,000 2 UNDER $75,000 7 REFUSED 8 DON'T KNOW 9 NO ANSWER Q:INCOME T: 1 "UNDER $15,000" 2 "$15,000 - $25,000" 3 "$25,000 - $35,000" 4 "$35,000 - $50,000" 5 "$50,000 - $75,000" 6 "$OVER 75,000"

Q:ENDIT T: Thank you, that is the end of the survey. We will be doing another short survey just on Y2K in January, may we call you back? 1 YES 2 NO --> SKIPTO INTOBS IF (ANS > 1) SKIPTO INTOBS Q:REINT T: Please give me your first name only. OPEN ENDED

243 TYPE RESPONDENT'S FIRST NAME ON ONE LINE TYPE EXACT RESPONSE BELOW Q:PHONNUM1 T: What is the phone number where we might most easily reach you? ENTER EXACT PHONE NUMBER INCLUDING AREA CODE - WITH NO HYPHEN 97 REFUSED 98 DON'T KNOW 99 NO ANSWER Q:PHONNUM2 T: Can you give me another number at which you can be reached? ENTER EXACT PHONE NUMBER INCLUDING AREA CODE - WITH NO HYPHEN 97 REFUSED 98 DON'T KNOW 99 NO ANSWER Q:INTOBS T: On behalf of the University of Oregon, Oregon State University and the Oregon State Library I would like to thank you for your time. INTERVIEWERS: PLEASE TYPE RELEVANT COMMENTS BELOW TYPE EXACT RESPONSE BELOW Q:NOQUAL T: NOTIFY SUPERVISOR

244

BIBLIOGRAPHY Alasuutari, Pertti. 1995. Researching Culture: Cultural Studies. Qualitative Method and

Altschull, J. Herber. 1990. From Milton to Mcluhan: The Ideas Behind American Journalism. White Plains, N.Y.: Longman. Aquinas, St. Thomas. 1272 (1966). Treatise on the Virtues. Translated by J. A. Oesterle. Notre Dame, IN: University of Notre Dame Press. ASNE. 1999. Leveraging Newspaper Assets [Web archive]. American Society of Newspaper Editors, Oct. 12 1998 [cited June 18 1999]. Assael, Henry. 1987. Consumer Behavior and Marketing Action. Third ed. Boston: PWS-Kent Publishing. Ball-Rokeach, Sandra J. 1985. The Origins of Individual Media-System Dependency: A Social Framework. Communication Research 12 (4):485-510. . 1998. A Theory of Media Power and a Theory of Media Use: Different Stories, Questions and Ways of Thinking. Mass Communication and Society 1 (1-2):5(40). Ball-Rokeach, Sandra J., and Melvin DeFleur. 1976. A Dependency Model of Mass Media Effects. Communication Research 3:3-21. Barnhurst, Kevin G., and Ellen Wartella. 1991. Newspapers and Citizenship: Young Adults' Subjective Experience of Newspapers. Critical Studies in Mass Communication 8 (June):195-209. Bartlett, John. 1968 (1882). Familiar Quotations. 14 ed. Boston: Little, Brown and Co. Bauer, Raymond A. 1963. The Initiative of the Audience. Journal of Advertising Research 3:2-7. Baukus, Robert A. 1999. Perception of Mediated Social Conflict: Media Dependency and Involvement. In Framing Friction, edited by M. S. Mander. Urbana: University of Illinois Press. Beam, Randal A. 1995. How Newspapers Use Readership Research. Newspaper Research Journal 16 (1):28(11).

245 . 1996. How Perceived Environmental Uncertainty Influences the Marketing Orientation of U.S. Daily Newspapers. Journalism & Mass Communication Quarterly 73 (2):295(19). Beatty, Sharon E., and Lynn R. Kahle. 1988. Alternative Hierarchies of the Attitude-Behavior Relationship: The Impact of Brand Commitment and Habit. Journal of the Academy of Marketing Science 16 (2):1-10. Becker, Lee B., and D. Charles Whitney. 1980. Effects of Media Dependencies: Audience Assessment of Government. Communication Research 7 (1):95-120. Bell, Catherine M. 1992. Ritual Theory, Ritual Practice. New York: Oxford University Press. Bentley, Clyde. 1998. 50 Years Later: "What It Means to Miss the Newspaper". Paper read at AEJMC Annual Conference, Aug. 5, at Baltimore, Maryland. . 1998. Building Media Brands. The Advertiser, September/October, 70(6). Berelson, Bernard. 1949. What "Missing the Newspaper" Means. In Communications Research, 1948-1949, edited by P. F. Lazarsfeld and F. N. Stanton. New York: Harper & Brothers. Biocca, Frank A. 1988. Opposing Conceptions of the Audience: The Active and Passive Hemispheres of Mass Communication Theory. In Communication Yearbook, edited by J. A. Anderson. Newbury Park: Sage. Blumler, Jay G. 1979. The Role of Theory in Uses and Gratifications Studies. Communication Research 6:9-36. Blumler, Jay G., and Elihu Katz. 1974. The Uses of Mass Communications: Current Perspectives on Gratifications Research. Beverly Hills, CA: Sage. Blumler, Jay. G., and Michael Gurevitch. 1996. Media Change and Social Change: Linkages and Junctures. In Mass Media and Society, edited by J. Curran and M. Gurevitch. London: Arnold. Bogart, Leo. 1981. Press and Public: Who Reads What, When, Where and Why in American Newspapers. Hillsdale, NJ: Lawrence Erlbaum Associates. . 1988. An Update on Readership. New York: Newspaper Advertising Bureau.

246 Brooks, Brian S., George Kennedy, Daryl R. Moen, and Don Ranly. 1999. News Reporting and Writing. Edited by P. Rossi. Sixth ed. Boston: Bedford/St Martin's. Bybee, Carl R. 1987. Uses and Gratification Research and the Study of Social Change. In Political Communication Research: Approaches, Studies, Assessments, edited by D. Paletz. Norwood, N.J.: Ablex. Carter, Jeffrey. 1999. Ritual: Perspectives and Dimensions. The Journal of Religion 79 (2):344(1). Chaffee, Steven H., and Jack M. McLeod. 1973. Individual Vs. Social Predictors of Information Seeking. Journalism Quarterly 50 (2):237-245. Chakravarty, Subrata N., and Carolyn Torcellini. 1989. Citizen Kane Meets Adam Smith. Forbes, Feb. 20, 82-85. CJR. 1999. Inside the Pulitzers. Columbia Journalism Review 38 (1):26. Collins, Randall. 1985. The Durkheimian Tradition. In Three Sociological Traditions. New York: Oxford University Press. Consoli, John. 1995. Selling Newspapers as Commodities. Editor & Publisher, APRIL 29, 1995, 11(1). Cooper, Lane. 1932. The Rhetoric of Aristotle. Englewood Cliffs, NJ: Prentice-Hall. Coppet, Daniel de. 1992. Understanding Rituals. Paper read at European Association of Social Anthropologists (Series), at London. Crompton, Susan. 1997. The Leisurely Pursuit of Reading. Canadian Social Trends Autumn (46):21(4). DeFleur, Melvin L., and Sandra Ball-Rokeach. 1989. Theories of Mass Communications. 5th ed. White Plains, NY: Longman. DeFleur, Milton L., and Everette E. Dennis. 1996. Understanding Mass Communication. 6th ed. Boston: Houghton Mifflin. Demers, David Pearce. 1996. The Menace of the Corporate Newspaper: Fact or Fiction? Ames, Iowa: Iowa State University Press. Dewey, John. 1922. Human Nature and Conduct; an Introduction to Social Psychology. New York,: H. Holt and Co. Durkheim, Emile. 1915. The Elementary Forms of Religious Life. New York: The Free Press. E&P. 1999. Read His Lips: "Circulation". Editor and Publisher, August 21, 20.

247 Ehrlich, Matthew C. 1996. Using "Ritual" to Study Journalism. Journal of Communication Inquiry 20 (2):3-17. Ewer, Cynthia. 2000. Habit, the Household Wonder Worker [Web page]. An Organized Home 1999 [cited Feb. 6 2000]. Available from http://organizedhome.com/organized/habit.html. Feder, Don. 1997. Good News & Bad News: The Trouble with Network News. The American Enterprise, Sept.-Oct., 44-45. Fink, Conrad. 1989. Lecture to the Media Management and Entrepreneurship Fellows of the Poynter Institute, St. Petersburg, Fla.,., Sept. 25, 1989. Fitzgerald, Mark. 1999. Growing Circulation, One Copy at a Time. Editor and Publisher, July 31, 12. . 1999. Poor Circulation: While Feeling Flush, Most Newspapers Watch Their Numbers Drop. Editor and Publisher, Nov. 13, 28. Fowler, Gilbert Len, Jr. 1985. An Examination of Readership Changes: Does Altering Publication Time Affect the Reading Habit? Newspaper Research Journal 7 (1):37-44. Garneau, George. 1994. Chopping Churn in Victoria, Texas; Steady Communication Helps Cut Subscriber Stops. Editor & Publisher, Dec. 10, 21(1). Gaziano, Cecile. 1990. Media Dependence for News: Some Neglected Groups. Mass Comm Review 17 (3):2(12). Giobbe, Dorothy. 1994. Newspaper Churn Levels Are as High as 70%. Editor & Publisher 127 (48):18(1). Gladney, George Albert. 1995. How Editors and Readers Rank and Rate the Importance of Eighteen Traditional Standards of Newspaper Excellence. Journalism and Mass Communications Quarterly 73 (2):319-331. Gollin, Albert E., and Paul A. Salisbury. 1980. Three Ways of Assessing Newspaper Readership Demographics. Newspaper Research Journal 1 (2):27-36. Hickey, Neil. 1998. Money Lust: How Pressure for Profit Is Perverting Journalism. Columbia Journalism Review, July/Aug, 28(8). Houston, Franklin S. 1986. The Marketing Concept: What It Is and What It Is Not. Journal of Marketing 50 (2):81(7). Hseih, Diana Mertz. 1997. Between Instinct and Habit: Aristotle on Habit. Honors Thesis, Washington University, St. Louis.

248 Imber-Black, Evan, and Janine Roberts. 1992. Rituals for Our Times : Celebrating, Healing, and Changing Our Lives and Our Relationships. 1st ed. New York: Harper Collins. IPA. 1999. Rules of Thumb. Des Plaines, IL: Inland Press Association. James, William. 1914. Habit. New York: H. Holt and Co. Janowitz, Morris. 1952. The Community Press in an Urban Setting. Second ed. Chicago: The University of Chicago Press. Kahle, Lynn. 1984. Attitudes and Social Adaptation: A Person-Situation Interaction Approach. Oxford: Pergamon. Kahle, Lynn R., and Sharon E. Beatty. 1987. The Task Situation and Habit in Attitude-Behavior Relationship: A Social Adaptation View. Journal of Social Behavior and Personality 2 (2):219-232. Kaplan, Justin. 1980. Walt Whitman, a Life. New York: Simon and Schuster. Katz, Elihu. 1959. Mass Communication Research and the Study of Culture. Studies in Public Communication 2:1(6). Katz, Elihu, Jay G. Blumler, and Michael Gurevitch. 1974. Utilization of Mass Communication by the Individual. In The Uses of Mass Communications: Current Perspectives on Gratifications Research, edited by J. G. Blumler and E. Katz. Beverly Hills: Sage. Kluckhorn, Clyde. 1942. Myths and Rituals: A General Theory. The Harvard Theological Review 35 (1):45(35). Krim, Seymour. 1988. The 215,000-Word Habit: Should I Give My Life to the Times? The Nation, April 23, 566(2). Lamb, Charles. 1833. Detached Thoughts on Books and Reading. In Bartlett's Familiar Quotations, edited by E. M. Beck. Boston: Little, Brown. Lasswell, Harold. 1948. The Structure and Function of Communications in Society. In The Communication of Ideas, edited by L. Bryson. New York,: Harper - Institute for Religious and Social Studies. Layton, Charles. 1999. What Do Readers Really Want? American Journalism Review, March, 46-63. Lindlof, Thomas R. 1995. Qualitative Communication Research Methods. Edited by J. G. Delia. 3 vols. Vol. 3, Current Communication: An Advanced Text Series. Thousand Oaks: London. Lippmann, Walter. 1922. Public Opinion. Free Press Paperback ed. New York: MacMillan.

249 Livingstone, Sonia. 1998. Relationships between Media and Audiences. In Media, Ritual and Identity, edited by T. Liebes and J. Curran. New York: Routledge. Loges, William E., and Sandra J. Ball-Rokeach. 1993. Dependency Relations and Newspaper Readership. Journalism Quarterly 70 (3):602-614. Lopez, Steve. 1997. Agate Addiction. Sports Illustrated, April 21, 102(1). Low, Elke. 1995. The High Expense of Wooing New Readers. Ideas, December/January, 24. Manheim, Jarol B. (1976) Can democracy survive television? Journal of Communications, 26:84-90, Spring 1976. 1976. Can Democracy Survive Television? Journal of Communications 26 (Spring):84-90. Marder, Eric. 1997. The Laws of Choice: Predicting Customer Behavior. New York: The Free Press. Martin, John R. 1995. Why Passengers Can't Ride Passenger Rail. Mass Transit 21 (6):66(4). Maslow, Abraham H. 1954. Motivation and Personality. 1st ed. New York: Harper. McQuail, Denis. 1994. Audiences: Theory and Research Traditions. In Mass Communication Theory: An Introduction: Sage. . 1998. With the Benefit of Hindsight: Reflections on Uses and Gratifications Research. In Approaches to Audiences: A Reader, edited by R. Dikinson, R. Haridranath and O. Linn. London: Arnold. McQuail, Denis, James Blumler, and J.R. Brown. 1972. The Television Audience, a Revised Perspective. In Sociology of Mass Communications, edited by D. McQuail. Harmondsworth: Penguin. Merskin, Debra. 1999. Media Dependency Theory: Origins and Directions. In Mass Media, Social Control and Social Change: A Macrosocial Perspective, edited by D. Demers and K. Viswanath. Ames, Iowa: Iowa State University Press. Miller, Susan H. 1989. Vanishing Readership: What Newspapers Need to Know About Women. Scripps Howard News, May, 4-9. Mott, Frank Luther. 1962. American Journalism. Third ed. New York: Macmillan. MPA. 2000. Single Copy and Subscription Circulation Per Issue for All ABC Magazines 1960-1998 [Web page]. Magazine Publishers of America 1999 [cited March 19 2000].

250 Mueller, Jennifer, and David Kamerer. 1995. Reader Preference for Electronic Newspapers. Newspaper Research Journal 16 (3):2-13. NAA. 1998. So Many Choices. So Little Time: Media Usage Study. Reston, VA: Newspaper Association of America. . 1998. Why Newspapers? [Web site]. Newspaper Association of America 1998 [cited Oct. 1 1998]. Available from http://www.naa.org/info/whynewspapers. . 1999. Facts About Newspaper 1998 [Web page]. Newspaper Association of America 1999 [cited March 7 1999]. Available from http://www.naa.org/info/facts.html. NAB. 1978. Identifying Prospects for the Newspaper: Frequent Readers, Infrequent Readers and Nonreaders. New York: Newspaper Advertising Bureau. . 1985. Newspapers in American News Habits: A Comparative Assessment. New York: Newspaper Advertising Bureau. . 1986. Key Facts About Newspapers and Advertising. New York: Newspaper Advertising Bureau. Nesbitt, Mary. 2000. The Readership Initiative: An Industrywide Collaboration to Grow Readership. Evanston, IL: Readership Institute. Newsweek. 2000. Newsweek Media Research Index [Website]. Newsweek Magazine 2000 [cited Feb. 10 2000]. Available from http://www.vmr.com/research/220.html. Nowak, Kjell. 1997. Effects No More? In Beyond Media Uses and Effects, edited by U. Carlsson. Gteborg, Sweden: Nordic Information Center for Media and Communication Research. Ogan, Christine L., Charlene J. Brown, and David H. Weaver. 1979. Characteristics of Managers of Selected U.S. Daily Newspapers. Journalism Quarterly 56 (2 (winter)):803-809. ONPA. 1999. Directory of the Oregon Newspaper Publishers Association. Portland: Oregon Newspaper Publishers Association. Oregon. 2000. Oregon Bluebook Online [Web page]. State of Oregon 2000 [cited March 21 2000]. Available from http://bluebook.state.or.us/. Pauly, John J. 1991. Interesting the Public: A Brief History of the Newsreading Movement. Communication 12:285-297.

251 Penrose, Jeanne, David H. Weaver, Richard R. Cole, and Donald Lewis Shaw. 1974. The Newspaper Nonreader 10 Years Later: A Partial Replication of Westley-Severin. Journalism Quarterly 51 (4):631638. Poindexter, Paula M. 1979. Daily Newspaper Non-Readers: Why They Don't Read. Journalism Quarterly 56 (Winter):764-770. Polansky, Sharon H., and Douglas W.W. Hughes. 1986. Managerial Innovation in Newspaper Organizations. Newspaper Research Journal 8 (1):1-12. Rarick, Galen R. 1973. Differences between Daily Newspaper Subscribers and Nonsubscribers. Journalism Quarterly 50 (2):265-270. Redmond, James, and Robert Trager. 1998. Balancing on the Wire: The Art of Managing Media Organizations. Boulder: Coursewise Publishing. Reimer, Bo. 1997. Texts, Contexts, Structures: Audience Studies and the Micro-Macro Link. In Beyond Media Uses and Effects, edited by U. Carlsson. Gteborg, Sweden: Nordic Information Center for Media and Communication Research. Rich, Carole. 1997. Writing and Reporting News: A Coaching Method. Second ed. Belmont: Wadsworth Publishing. Rieder, Rem. 1999. Playing to Their Strengths. American Journalism Review 21 (3):6(1). Robertson, Thomas S. 1970. Consumer Behavior. Glenview, Ill: Foresman. Rockmore, Milt. 1995. How Do You Create a Newspaper Brand? Editor & Publisher, Oct. 14, 16(2). Rosengren, Karl Erik. 1974. Uses and Gratifications: A Paradigm Outlined. In The Uses of Mass Communications: Current Perspectives on Gratifications Research, edited by J. G. Blumler and E. Katz. Beverly Hills: Sage. Rothenbuhler, Eric W. 1998. Ritual Communication : From Everyday Conversation to Mediated Ceremony. Thousand Oaks: Sage Publications. Rouner, Donna. 1984. Active Television Viewing and the Cultivation Hypothesis. Journalism Quarterly 61:168-173. Rubin, Alan M. 1993. Audience Activity and Media Use. Communication Monographs 60:98-105.

252 Rubin, Alan M., and Sven Windahl. 1986. The Uses and Dependency Model of Mass Communication. Critical Studies in Mass Communication 3:184-199. Rykken, Rolf. 1989. Readership Decline Brings Newspapers to Crossroads. Presstime, March, 22-24. Salant, Pricilla, and Don A. Dillman. 1994. How to Conduct You Own Survey. New York: John Wiley and Sons. Saltzman, Joe. 1995. Its Habit-Forming. USA Today Magazine, May, 39(1). Schrder, Kim. 1987. Covergence of Antagonistic Traditions? The Case of Audience Research. European Journal of Communication 2:7-31. Schweitzer, John C. 1977. Life Style and Readership: A Review of the Research Literature. In ANPA News Research Report. Washington, D.C.: American Newspaper Publishers Association. Sobal, Jeff, and Marilyn Jackson-Beeck. 1981. Newspaper Nonreaders: A National Profile. Journalism Quarterly 58 (1):9(5). Sohn, David. 1975. A Nation of Vidiots. New York: Simon and Schuster. Stempel III, Guido H. , and Thomas Hargrove. 1996. Mass Media Audiences in a Changing Media Environment. Journalism and Mass Communications Quarterly 73 (3):549-558. Stone, Gerald. 1987. Examining Newspapers: What Research Reveals About America's Newspapers. Vol. 20, The Sage Commtext Series. Newbury Park, CA: Sage. Stone, Gerald, and Timothy Boudreau. 1995. 1985, 1994: Comparisons of Reader Content Preferences. Newspaper Research Journal 16 (4):1328. Stone, Gerald C. 1986. No Rest-in-Peace for Readers after PM's Demise`. Newspaper Research Journal 8 (1):13-27. Stone, Gerald C., and Roger V. Wetherington, Jr. 1979. Confirming the Newspaper Reading Habit. Journalism Quarterly 56 (Autumn):554 (9). Stone, Gerald, and Donna Besser Stone. 1990. Lurking in the Literature: Another Look at Media Use Habits. Mass Comm Review 17 (12):25(10). Stone, Gerald, and J. Windhause. 1983. Young Adults' Content Preferences Based on Morning Vs. Evening Reading Time. Newspaper Research Journal 4 (2):9.

253 Taylor, Donald. 2000. Comments on Uses and Dependency, March 2. Taylor, Donald S.E. 1991. Application of the Uses and Dependency Model of Mass Communication to Development Communication in the Wester Area of Sierra Leone. Dissertation, School of Commication Studies, Kent State University, Kent, Ohio. Tipton, Leonard. 1978. ANPA Newspaper Readership Studies. In ANPA News Research Report. Washington, D.C.: American Newspaper Publishers Association. Tocqueville, Alexis de. 1969. Democracy in America. Translated by G. Lawrence. Edited by J. P. Mayer. Garden City, N.Y.: Anchor Books. Tull, Donald S., and Lynn R. Kahle. 1990. Marketing Management, Macmillan Series in Marketing. New York: Macmillan. U.S. Census Bureau. 1987. Statistical Abstract of the United States: 1988. 108th ed. Washington, D.C.: U.S. Government Printing Office. . 1998. Statistical Abstract of the United States. Washington, DC: U.S. Government Printing Office. UTSA. 2000. Welcome to Factor Analysis [On-line text]. University of Texas at San Antonio Sociology Department, Jan. 1, 1999 2000 [cited April 18 2000]. Available from www.csbs.utsa.edu/social&policy/soc/masters/factoranal.htm. Waksberg, J. 1978. Sampling Method for Random Digit Dialing. Journal of the American Statistical Association 73:40-66. Walker, Leslie. 2000. All the News That Fits on a Screen. Washington Post, Feb. 2, EO1. Wallace, Anthony F.C. 1966. Religion: An Anthropological View. New York: Random House. WAN. 2000. About Newspapers in Education [Web site]. World Association of Newspapers 2000 [cited March 2 2000]. Available from http://www.wan-press.org/nie/#background. Wanta, Wayne, Yu-Wei Hu, and Yi-Chen Wu. 1995. Getting More People to Read More Newspapers: Factors Affecting Newspaper Reading. Newspaper Research Journal 16 (1):103(13). Weinstein, Art. 1987. Market Segmentation. Chicago: Probus Publishing. Westley, Bruce, and Werner J. Severin. 1964. A Profile of the Daily Newspaper Non-Reader. Journalism Quarterly 41:45-50.

254 Wilcox, Delos F. 1900. The American Newspaper: A Study in Social Psychology. Annals of the American Academy of Political and Social Science 16:56-92. Willis, Jim. 1998. Class Appeal Vs. Mass Appeal in Newspapers. The American Editor, March, 8(22).