You are on page 1of 13

Measuring the Involvement Construct

Author(s): Judith Lynne Zaichkowsky

Source: The Journal of Consumer Research, Vol. 12, No. 3 (Dec., 1985), pp. 341-352
Published by: The University of Chicago Press
Stable URL:
Accessed: 18/04/2009 08:43

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.

JSTOR is a not-for-profit organization founded in 1995 to build trusted digital archives for scholarship. We work with the
scholarly community to preserve their work and the materials they rely upon, and to build a common research platform that
promotes the discovery and use of these resources. For more information about JSTOR, please contact

The University of Chicago Press is collaborating with JSTOR to digitize, preserve and extend access to The
Journal of Consumer Research.
Measuring the Involvement Construct*

A bipolar adjective scale, the Personal Involvement Inventory (Pll), was developed
to capture the concept of involvement for products. The scale successfully met
standards for internalreliability,reliabilityover time, content validity,criterion-related
validity, and construct validity. Tests of construct validity demonstrated that the
scores were positively related to perceived differences among brands, brand pref-
erences, interest in gathering informationabout the product category, and comparison
of product attributes among brands.

R esearchersof consumerbehaviorhave historically to the different applications of the term "involvement."

developed a number of complex theories in the The literature suggests that a person can be involved
attempt to explain and predict the behavior of the con- with advertisements (Krugman 1962, 1965, 1967,
sumer (e.g., Bettman 1979; Engel, Kollat, and Blackwell 1977), with products (Howard and Sheth 1969; Hupfer
1978; Howard and Sheth 1969). These theories propose and Gardner 1971), or with purchase decisions (Clarke
that consumers actively search for and use information and Belk 1978). Involvement with these different objects
to make informed choices. This implies that the con- leads to different responses. For example, involvement
sumer is an intelligent, rational, thinking, and problem- with ads leads one to give more counterarguments to
solving organism, who stores and evaluates sensory in- the ad (Wright 1974). Involvement with products has
puts to make a reasoned decision (Markin and Naray- been hypothesized to lead to greater perception of at-
ana 1975). tribute differences, perception of greater product im-
However, a great deal of consumer behavior does not portance, and greater commitment to brand choice
involve extensive search for information or a compre- (Howard and Sheth 1969). Involvement with purchases
hensive evaluation of the choice alternatives, even for leads one to search for more information and spend
the purchase of major items (Olshavsky and Granbois more time searching for the right selection (Clarke and
1979). The average consumer makes dozens of mun- Belk 1978). Therefore, each area might have its own
dane decisions each day, few of which may be of im- idiosyncratic result of the state of being involved with
portance. For such decisions, it may be inappropriate the object.
to assume an active information processor (Kassarjian Researchers generally use the resulting behaviors as
1978, 1981). This idea has led theorists to view con- indicators of the level of involvement. Previous research
sumer behavior in terms of a two-fold dichotomy: low has examined involvement with advertisements via a
involvement consumer behavior and high involvement five-point scale that measures the degree of attention
consumer behavior (Engel and Blackwell 1982). to the ad (Wright 1973, 1974). Involvement with prod-
ucts has been measured by several methods: rank-or-
THE PROBLEM dering products (Sheth and Venkatesen 1968), rating a
Although researchers agree that the study of low ver- series of products on an eight-point concentric scale as
sus high involvement states is interesting and important, to their importance in the subject's life (Hupfer and
there is currently little agreement about how to best Gardner 1971), asking how important it is to get a par-
define, and hence measure, the construct of involve- ticular brand (Cohen and Goldberg 1970), or finding
ment (Cohen 1983). The reasons for the diverse defi- the total times that subjects report "don't know" for a
nitions and measures of involvement are perhaps due series of brands (Ray 1973). On a broader level, in-
volvement has been measured by administering Likert
*This article was a finalist in the 1984 Robert Ferber Award for statements that were thought to tap the underlying con-
Consumer Research competition for the best interdisciplinary article cept-e.g., the product means a lot to me, it matters to
based on a recent doctoral dissertation. The award is cosponsored by me, or the product is important to me (Lastovicka and
the Association for Consumer Research and the Journal of Consumer Gardner 1978a; Traylor 1981).
`-Judith Lynne Zaichkowsky is Assistant Professor of Marketing,
These diverse measures pose several problems for re-
Faculty of Business Administration, Simon Fraser University, Bur- searchers. If conflicting results are obtained, we do not
naby, B.C., Canada V5A 1S6. The author wishes to thank Hal Kas- know if the discrepancy is due to different measures or
sarjian, who chaired the dissertation on which this article is based at to different behaviors. Second, many scales are single-
the University of California, Los Angeles, and the Canada Council item measures and may not capture the total involve-
for financial support. A special thanks goes to The American Uni-
versity and Nanette Brown for help in word processing the manuscript. ment concept. Finally, single-item measures have low
reliability, and current multiple-item measures have not
(?)JOURNAL OF CONSUMER RESEARCH* Vol. 12 * December 1985

been tested for internal reliability, stability, or validity. goods did not seem to apply to durable goods and vice
Hence a standardized, general, valid, and multiple-item versa.
measure of involvement should be useful. The most effective and generalizable type of scale ap-
peared to be a semantic differential type (Osgood, Suci,
and Tannenbaum 1957). The Semantic Differential
BACKGROUND AND CRITERIA consists of a series of bipolar items, each measured on
FOR MEASURING INVOLVEMENT a seven-point rating scale. It is easy to administer and
score, takes only a few minutes to complete, and is ap-
A measure of involvement-independent of the be- plicable to a wide array of objects. The descriptors or
havior that results from involvement-would allow the
phrases easily relate across product categories and can
researcher to use the same measure across various re-
be appropriate to other domains, such as purchase de-
search studies. This measure should also be sensitive to
cisions or advertisements. (However, the main focus of
the proposed areas that affect a person's involvement
this article and scale development is involvement with
level. These areas might be classified into three cate-
products.) The steps taken to develop the measure were:
gories (Bloch and Richins 1983; Houston and Roths-
child 1978): 1. Define the constructto be measured.
2. Generateitems that pertainto the construct.
1. Personal-inherent interests, values, or needs that
motivate one towardthe object 3. Judge the content validity of generateditems (item
2. Physical-characteristicsof the objectthat cause dif-
ferentiationand increaseinterest 4. Determinethe internal reliabilityof items judged to
have content validity (item reduction).
3. Situational-something that temporarily increases
relevanceor interesttowardthe object 5. Determine the stability of internally reliable items
over time (item reduction).
In Houston and Rothschild's (1978) framework, differ- 6. Measurethe content validityof the 20 selecteditems
ent situations and different people are two factors that as a whole.
lead to various levels of involvement. Houston and
Rothschild integrate physical characteristics of the 7. Measurethe criterion-relatedvalidity, which is the
product as part of the situational factor. Coinciding with ability of the scale to discriminateamong different
Bloch and Richins (1983), the present article separates productsfor the same people and differentsituations
for the same productand same people.
the physical from the situational and allows the same
physical object to be subjected to different levels of in- 8. Test the constructvalidityor theoreticalvalue of the
volvement given different situations. scale by gatheringdata and testing whetherthe scale
The evidence for the three factors-physical, per- discriminateson self-reportedbehavior.
sonal, or situational-that influence the consumer's
level of involvement or response to products, advertis- DEFINING THE CONSTRUCT
ing, and purchase decisions is found in the literature.
For example, Wright (1974) found that variation in the This article will adopt the general view of involve-
type of media-print versus audio-influenced the re- ment that focuses on personal relevance (Greenwald
sponse given to the same message (physical). Lastovicka and Leavitt 1984; Krugman 1967; Mitchell 1979;
and Gardner (1 978a) demonstrated that the same prod- Rothschild 1984). In the advertising domain, involve-
uct has different involvement levels across people (per- ment is manipulated by making the ad "relevant:" the
sonal), and Clarke and Belk (1978) demonstrated that receiver is personally affected, and hence motivated, to
different purchase situations for the same products respond to the ad (e.g., Petty and Cacciopo 1981). In
cause differences in search and evaluation or raise the product class research, the concern is with the relevance
level of involvement (situational). Based on this prior of the product to the needs and values of the consumer.
reasoning, a measure of involvement might be devel- In purchase decision research, the concern is that the
oped that would pick up differences across people, ob- decision is relevant, and hence that the consumer will
jects, and situations. be motivated to make a careful purchase decision (e.g.,
Different types of scales were pretested before select- Clarke and Belk 1978). Although each is a different do-
ing a measurement approach that seemed to be gener- main of research, in general, high involvement means
alizable across all product categories. First, a series of personal relevance (Greenwald and Leavitt 1984).
vignettes was developed to represent involvement. The In this study, the definition of involvement used for
vignettes were similar to scenarios found in Lastovicka the purposes of scale development was:
and Gardner (1978b). Problems arose with developing A person's perceived relevance of the object based on
enough generalizable scenarios for a reliable scale. Lik- inherentneeds, values, and interests.
ert scale items proposed a problem because items that This definition recognized past definitions of involve-
seemed to be appropriate for frequently purchased ment (e.g., Engel and Blackwell 1982; Krugman 1967;

Mitchell 1979). Judging from previous writings, this were generally not negative-as they would be if mea-
definition may be applied to advertisements, products, suring attitudes-but rather were "who cares" descrip-
or purchase decisions. Early work by Krugman (1962, tors, e.g., unimportant, unexciting, doesn't matter, or
1967) in advertising focused on personal connections. of no concern.
Wright (1974) defined involvement with advertising as Five new judges then rated the remaining 43 word
the receiver's perception of the relevancy of the ad con- pairs using the same procedure. Only 23 items were
tent to some pending problem. In the area of product consistently rated as representing the involvement con-
class involvement, Howard and Sheth (1969) used the struct (80 percent agreement over products, purchase
terms importance of purchase and involvement inter- decisions, and advertisements for each word pair). This
changeably; they defined involvement in terms of a meant that at least 12 of the possible 15 judgments for
person's needs or values. Hupfer and Gardner (1971) each word pair (five judges over three objects) had to
defined involvement as a general level of interest in or be rated as representative of the involvement construct.
concern about an issue without reference to a specific Agreement across judges and within each area for the
position. Finally, Houston and Rothschild (1978) re- 23 word pairs was as follows: advertisements, 84 per-
ferred to response involvement and defined it as a func- cent; products, 87 percent; and purchase decisions, 77
tion of enduring involvement or a need derived from a percent.
value in the individual's hierarchy of needs. Twenty-three was assumed to be too low a number
of items with which to start data collection (French and
Michael 1966; Nunnally 1978). Thus, seven additional
ITEM GENERATION AND items were added to the item pool to raise the initial
CONTENT VALIDITY number to 30 (five of these seven were eventually
dropped). For example, trivial-grand (45 percent
A semantic differential scale was to be developed agreement) was changed to trivial-fundamental, and
based on the earlier definition of involvement. Thus, a inspiring-discouraging (55 percent agreement) was
list of 168 word pairs was generated to represent this changed to inspiring-uninspiring and returned to the
concept of involvement. Examples of those pairs are list. Therefore, a thirty-item scale emerged from the
important-unimportant, interested-uninterested, and content validity phase that trained and knowledgeable
exciting-unexciting. The first step was to judge the pro- judges agreed measured involvement over three do-
posed items for content validity-how well the chosen mains: products, advertisements, and purchase deci-
items represent the defined concept. Content validity sions. However, this study focused on, and further val-
of the 168 word pairs was tested in two phases: (1) initial idation procedures were carried out on, involvement
deletion of poor word pairs, and (2) finer judging of the with products.
more appropriate word pairs.
Three expert judges (senior Ph.D. candidates in con-
sumer behavior) were given this study's definition of INTERNAL SCALE RELIABILITY
involvement and instructed to rate the 168 word pairs
three times: first, replacing the word "object" in the The next task was to administer the 30 items as a
definition with "product;" second, replacing the word scale over different product categories to measure the
"object" with "advertisement;" and third, replacing the internal consistency or inter-item correlation. Two
word "object" with "purchase decision." Each word product classes-watches and athletic shoes-were se-
pair was rated on the following scale: (1) clearly rep- lected because they were thought to be used by the sub-
resentative of involvement, (2) somewhat representative jects. One hundred and fifty-two undergraduate psy-
of involvement, and (3) not representative of involve- chology students completed the scale during class time.
ment. Word pairs that were not rated as representative Approximately half of the subjects filled out the scale
of involvement for any advertisement, purchase deci- pertaining to athletic shoes and the other half filled out
sion, or product were dropped. Examples of deleted the scale pertaining to watches. The results show that
word pairs are adequate-inadequate, controversial- for both product categories, 26 bipolar items had an
noncontroversial, and naive-sophisticated. item-to-total score correlation of 0.50 or more, and a
Word pairs that were dropped included traditional Cronbach alpha level of 0.95.
measures of attitudes used in the psychology and mar- Six adjective pairs with relatively low item-to-total
keting literature. Word pairs such as good-bad, pleas- correlations were dropped; interestingly, most of these
ant-unpleasant, nice-awful, and like-dislike (e.g., adjective pairs had been returned earlier to the item
Loken 1984; Mitchell and Olson 1981) were judged to pool. Factor analyses, using varimax rotation with
be unrepresentative of involvement. The judges decided squared multiple correlations in the diagnonals for fac-
that other word pairs, such as valuable-worthless and tor extraction, were carried out over both products to
appealing-unappealing would remain, as they seemed check if the items selected for deletion loaded onto one
to measure involvement. Items at the low end of the particular dimension or were amorphous across factors.
bipolar scale that represent the low end of involvement For both watches and athletic shoes, one factor ex-

plained the major variation in the data, accounting for fits neatly on one page and only takes a few moments
70.3 percent and 69.3 percent of the common variance, to complete. The scale was then counterbalanced so
respectively (eigenvalues 13.3 and 13.2). Watches had that ten random items were reverse scored. Since each
two more factors, accounting for 11.6 percent and 5.6 bipolar item was rated on a seven-point scale, the total
percent of the common variance (eigenvalues 2.2 and possible score ranged from a low of 20 to a high of 140.
1.1), and athletic shoes had three more factors, ac- The scale was named the Personal Involvement Inven-
counting for 11.7 percent, 5.9 percent, and 5.7 percent tory (Pll) and is listed in Appendix A.
of the common variance (eigenvalues 2.2, 1.2, and 1. 1).
The results of the factor analyses showed that the SECOND CONTENT VALIDITY
items selected for deletion did not load together on any A second measure of content validity was obtained
unique factor across either product category. Since the from the open-ended responses of 45 MBA students
first factor accounts for approximately 70 percent of over three product categories: 35mm cameras, red wine,
the variance, and none of the remaining items had a and breakfast cereals. After completing the scales for
loading of zero or less on that first dimension, the scale each product, subjects answered the following open-
development continued on the assumption of a simple ended question:
linear combination of the individual items (Comrey
1973). The assumption is that no individual item is suf- Now we would like you to state, in your own words, why
ficient, and that it is the scale taken as a whole that you ratedeach productcategoryas you did.
tends to measure the involvement construct (Nunnally Subjects were then divided into three groups-high,
1978). medium, or low-for each product class according to
their scale scores.2 Examples of the open-ended respon-
TEST-RETEST RELIABILITY ses appear in the Exhibit.
Two expert judges (senior Ph.D. candidates in con-
Test-retest reliability of the remaining 24 items was sumer behavior) blind to the scale scores evaluated the
examined over two new subject samples and four new total set of open-ended responses. For each product
product categories. Sixty-eight psychology students ini- category, the judges sorted the comments into three
tially rated calculators and mouthwash. Forty-five MBA groups indicative of low involvement, medium in-
students rated breakfast cereals and red wine. The order volvement, and high involvement with the product cat-
of the products was counterbalanced-half of the sub-
egory, based on how well the responses represented in-
jects in each group rated one product category first, and volvement, as defined earlier.
the other half rated the other product category first. Interjudge reliability on the classification of the re-
The scales were administered during class time and took
sponses was 80 percent agreement for 35mm cameras,
about five minutes to complete.
84 percent agreement for red wines, and 80 percent
Three weeks later the scales were administered over
agreement for breakfast cereals. Classifications on which
the same product categories to the same subjects. Thir-
the two expert judges did not agree were then given to
teen psychology students and 19 MBA subjects were
lost to attrition; thus, 55 psychology students and 26
MBA students were used to measure test-retest reli- of items would be almost as reliable as the 20 items. The problem of
reducing the scale to fewer items lies in deciding which items to select
ability. The average Pearson correlation between Time as subsets, since individual items differed in their reliability across
I and Time 2 on the 24 items was 0.90. Individual item- product categories. A subset of items that may approach the reliability
to-item correlations ranged from 0.31 to 0.93. Four ad- of the 20 items for one product may not approach the same reliability
ditional items with average test-retest correlations below for another product. This variation is evident in that the test-retest
0.60 were deleted. The resulting twenty-item involve- total score correlation ranged from 0.88 to 0.93 over products, and
test-retest for the 20 individual items ranged from 0.44 to 0.93 over
ment score test-retest correlations for each product were various products. The twenty-item measure should outperform any
as follows: calculators, r = 0.88; mouthwash, r = 0.89; subset of the scale; besides, decreasing the number of items would
breakfast cereals, r = 0.88; and red wine, r = 0.93. These not really make the scale any easier to administer, but may serve to
product categories were also tested for internal scale decrease the domain of items judged as being representative of in-
volvement and also lower the reliability of the scale. Researchers who
reliability. The Cronbach alpha ranged from 0.95 to may use this scale are warned not to haphazardly reduce the number
0.97 over the four products. of items.
Therefore, a twenty-item scale emerged from the in- 2The classification of subjects into low, medium, and high scores
ternal reliability and stability phases of scale develop- was based on an overall distribution developed over 13 product cat-
ment for products. Twenty items allowed an adequate egories (Table 3) and several hundred subjects. All scores were tab-
ulated on the PII scale range presented in the Figure. Subjects whose
sampling of the possible items that represent involve- PII scores fell into the bottom 25 percent of the overall distribution
ment with products and yet was long enough to ensure were classified as having low involvement with the product. Subjects
a high level of reliability.' On a practical level, the scale whose PII scores fell into the middle 50 percent of the distribution
were classified as having medium involvement, and subjects whose
PII scores were in the top 25 percent of the distribution were classified
'Although the current analyses do not suggest what the reliability as having high involvement with the product. For development of
is for subsets of the scale items, the case may be that a smaller number this classification scheme see Appendix B.



35mm Cameras
Collapsed for
1. High involvement for cameras (score greater than 110). Judges' ratings Chi-square
a. Subject 1. Cameras are important, but not essential. They Scale
provide a creative and historical outlet for me. scores Low Medium High (Total) Low Medium High
b. Subject 12. Cameras interest me and are an important hobby
to me. 35 mm Camerasa
2. Low involvement for cameras (scores less than 70). Low 7 1 0 (8) 11 13 7
a. Subject 17. Because I never use 35mm cameras and am not Medium 4 12 7 (23) 0 4 10
extremely interested in them. High 0 4 10 (14)
b. Subject 37. It's a nice product to have but not a high priority.
(Total) (11) (17) (17) (45)
I have several but as I recall, none of the purchases was an
"involved" purchase.
Red wineb
Red Wine Low 12 1 0 (13) 12 1 0
Medium 8 9 8 (25) 8 10 14
1. High involvement for red wine (score greater than 1 10).
High 0 1 6 (7)
a. Subject 22. Red wine adds a lot to the appropriate meals.
b. Subject 6. I have always wanted to know more about wines (Total) (20) (1 1) (14) (45)
and enjoy it when people I know teach me about them.
Breakfast cerealsc
2. Low involvement for red wine (score less than 70).
a. Subject 20. I'm not interested in wines nor do I particularly Low 19 3 0 (22) 19 3 0
appreciate the mystique that surrounds wines, in general. Medium 9 9 1 (19) 9 11 3
b. Subject 36. OK for socials and getting drunk. High 0 2 2 (4)
(Total) (28) (14) (3) (45)
Breakfast cereals
1. High involvement for breakfast cereals (score greater than 1 10). x2= 10.4, df = 2, p < 0.01.
b 2= 17.0, df = 2, p < 0.001.
a. Subject 27. I eat cereal, healthy efficient "wake up America." C 2 = 11.2, df = 2, p < 0.01.

Cereal is good for you. NOTE: As more than 20 percent of the expected cell frequencies dropped below 5, either
b. Subject 8. Because they are diet foods. the low or high row was collapsed into the medium row to compute the statistic.

2. Low involvement for breakfast cereals (score less than 70).

a. Subject 3. I think breakfast cereals are a sham. I only eat having either high or low involvement (Bowen and
grapenuts. It infuriates me to see breakfast cereals Chaffee 1974; Hupfer and Gardner 1971; Lastovicka
advertised to be eaten with toast, juice, etc. What's the use, and Gardner 1978a; Traylor 1981) were presented to a
jaw exercise? I refuse to buy cereal for my child. group of 68 undergraduate psychology students. As in
b. Subject 31. I eat cereal for convenience; it is easy and fast. I
have no interest in them nor am I fascinated with them. Hupfer and Gardner (1971), subjects rated each product
on an eight-point scale: extremely unimportant in my
life (1) to extremely important in my life (8).
a third judge to classify. The categories of responses, as From these 21 products, four were selected for mea-
grouped by the scale scores, were compared to the cat- surement: bubble bath (mean (X) = 2.35); facial tissue
egories of responses as grouped by the expert judges. (X = 5.25); jeans (X = 6.6); and automobiles (X = 7.9).
The results are presented in Table 1. These data indicate Bubble bath was previously selected as a low involve-
a significant relationship between the scale scores and ment product, and jeans a high involvement product
the open-ended responses from the subjects, thus adding by Clarke and Belk (1978). Facial tissue was previously
an additional modicum of support to the validity of the identified as a low involvement product, and automo-
scale. biles as a high involvement product by Lastovicka and
Gardner (1978a).
CRITERION-RELATED VALIDITY The twenty-item involvement scale (Pll) was admin-
istered for each of the four product categories to a fresh
Criterion-related validity is demonstrated by com- sample of 47 undergraduate psychology students during
paring the scores from the developed instrument with class time. The PII mean scores and standard deviations
one or more external variables that provide a direct for each product were as follows: bubble bath X = 69,
measure of the characteristic in question (French and s = 38 (Males X = 55, Females X = 74); facial tissue X
Michael 1966). The external variable selected as a cri- = 87, s = 26; jeans X = 99, s = 21; and automobiles X
terion was the simple ordering or classification of prod- = 122, s = 19.
ucts into low or high involvement categories. A repeated measures analyses of variance showed an
Twenty-one products classified in other studies as overall significant difference among the product means

(F(3, 138) = 39.9, p < 0.001). Furthermore, each mean males and 50 females) at a major university.Subjects
was found to be significantly different from each of the werepersonallycontactedat workby the researcherand
others (p < 0.01). These results are in agreement with askedif they would participate.Those who agreedwere
previous studies that have stated that facial tissue and given a questionnaireand asked to complete it at their
bubble bath have lower involvement levels than jeans office desk. The researcherthen returnedin about an
and automobiles. hour to collect the questionnaires from the subjects.
Seven additional subjects agreed to fill out the ques-
CONSTRUCT VALIDITY tionnairebut nevercompletedit. The medianage range
of the subjects was 35-44 years and the median edu-
Studies of construct validity check the theory under- cation level was some college. The productsselectedfor
lying the test (French and Michael 1966). Three steps evaluation were instant coffee, laundrydetergent,and
are involved in construct validity. First, from the in- colortelevision.3Thesewerechosento representa range
volvement literature, propositions are made about the of productsthought to be used by the subjects.
behavior of people with high and low scores. Second, Subjectswereclassifiedinto threegroupsto compare
data is gathered to test if the scale discriminates on be- the responsesbetween subjects who had low involve-
havior, and third, an inference is made as to whether ment with the productcategoryand those who had high
the theory is adequate to explain the data collected. involvement with the product category. This classifi-
cation scheme is the same as that found in the second
content validity section and furtherexplained in Ap-
TheoreticalPropositions of Involvement pendix B. The particularquestion of interestwas:"Did
Various propositions about differences in low and subjectshaving low PII scores for the productcategory
high involvement behavior were selected after reviewing respondto the five statementsdifferentlythan did those
theoretical papers by several authors (e.g., Belk 1981; having high PII scores for the product category?"
Bowen and Chaffee 1974; Lastovicka 1979; Lastovicka Planned comparisons, by simple t-tests, were carried
and Gardner 1978b; Mitchell 1979; Robertson 1976; out between the low and high PII scoresfor each state-
Tyebjee 1979). Generally, there seems to be some ment and product category.Before comparingthe dif-
agreement on what constitutes the differences between ferences between the low and high group in their re-
having high or low involvement in a product class. Un- sponsesto the statementswithin each productcategory,
der the low involvement condition, researchers propose: a one-wayMANOVAwas computedoverthe five state-
ments for each product category to determine if the
1. A relative lack of active information seeking about overallpatternof responsesacrossthe five propositions
brands was significant.
2. Little comparisonamong productattributes In addition, the Pearson correlations between the
scale score (range= 120) and the responsesto the state-
3. Perceptionof similarityamong differentbrands ments (range= 7) werecomputed.The cell means, cor-
4. No special preferencefor a particularbrand relations,and resultsof t-testsbetweenthe low and high
cells are presentedand summarizedin Table 2.
Based on these various theoretical propositions, the
following specific statements were developed and then Results. The Cronbach alpha, the mean, and the
administered to subjects with the Pll over various standarddeviationof the Pll scoresfor the threeproduct
product categories: categorieswere as follows:
1. I would be interestedin reading information about Cronbachalpha Mean (s)
how the productis made.
Instantcoffee .97 66 (40)
2. I wouldbe interestedin readingthe ConsumerReports Colortelevision .99 97 (30)
articleabout this productcategory. Laundrydetergent .97 103 (23)
3. I have compared product characteristics among
brandsof this product. These scoresprovidetwo unexpectedresults:first,a rel-
4. I think there are a great deal of differencesamong
atively high PII score for laundrydetergent(103), and
brandsof this product. second, a relatively low Pll score for color television
(97). These resultsshould be interpretedin the context
5. I have a most preferredbrandof this product. of the sample population used in this study: this rela-
These statements were rated on a seven-point scale from tively homogeneousgroupof middle-agedfemalesmay
have viewed laundrydetergentas more involving than
strongly disagree (1) to strongly agree (7).
Method. The PII and the specific questions related
to involvement were administered over three products 3PIIscales for the productcategoryof red wine and two purchase
to 28 clerical and 29 administrative staff members (7 situationsfor red wine were also includedfor other purposes.


Instant coffee Laundry detergent Color television

Construct validity Low Medium High Low Medium High Low Medium High
statementa (32) (12) (12) rb (4) (28) (25) r (9) (26) (12) r

1. I would be interested in 3.28 4.42 4.25 .30c 1.25e 4.04 4.48 .37c 3.56 4.00 4.23 .14
reading information (2.0) (2.3) (2.3) (5) (1.7) (2.4) (2.1) (1.9) (2.1)
about how the product is
2. 1 would be interested in 3.00e 4.75 4.92 47c 2.75 f 4.46 5.00 33c 4.56 4.65 5.36 .27c
reading the Consumer (1.8) (2.3) (2.3) (2.9) (2.0) (2.1) (1.9) (1.9) (2.1)
Reports article about this
3. I have compared product 2.59e 3.42 5.25 .52c 1.75 4.36 4.80 .42c 3.11 e 3.85 4.59 *23d
characteristics among (1.8) (2.1) (2.0) (1.5) (1.8) (2.4) (1.9) (1.9) (2.3)
4. I think there are a great 3.94e 4.67 6.33 .63c 2.25e 4.00 5.20 .42c 4.11 e 4.85 5.73 .33c
deal of differences (1.6) (1.1) (.8) (1.0) (1.7) (2.1) (1.2) (1.5) (1.7)
among brands.
5. 1have a most-preferred 2.88e 4.83 6.17 .68c 2.50e 4.68 5.44 .42c 2.56e 4.77 5.55 .50c
brand of this product. (1.9) (1.8) (1.7) (3.0) (1.6) (1.9) (1.4) (1.7) (1.9)

The construct validity statements are measured on a seven-point scale: (1) strongly disagree to (7) strongly agree.
br = Pearson Correlation between Pll score and response to construct validity question.
Cp < 0.01.
dp < 0.05.
e Low scores significantly different than high scores p < 0.01.
f Low scores significantly different than high scores
p < 0.05.
NOTE: Numbers in parentheses in Table heading are numbers of subjects in each group. Numbers in parentheses in Table body are standard deviations.

color television because it may be their responsibility level of involvement as a mediating variable in infor-
to do the family laundry. If this is true, they would mation search. Thus, high scale scorers should indicate
value the product's benefits and they would be likely more interest in product information than low scorers.
to be interested in the quality of the product because Subjects were given two statements pertaining to in-
they need the product to perform their household duties. formation search over the three product categories. The
Televisions, however, may not fall under their respon- first statement was "I would be interested in reading
sibility for maintenance or interest them as much. Elec- information about how the product is made (instant
tronics, solid state, or color tuning may not be relevant coffee and laundry detergent) or works (color televi-
to them. And if a television does not affect them per- sion)." The second statement was "I would be interested
sonally, housewives might have relatively low involve- in reading the Consumer Reports article about
ment with this product. The results of the analyses of the information search
The results of the MANOVA for all five statements questions are generally in agreement with the theory of
were significant for the three products (instant coffee involvement. High scorers tended to be more interested
F(10, 98) = 6.56,p < 0.001; laundry detergent F(l0, 100) in information pertaining to the product than low scor-
= 2.34, p < 0.05; and color television F(10, 100) = 2.00, ers. All correlation coefficients were significantly dif-
p < 0.05). This indicated that there were significant dif- ferent from zero, with the exception of the product cat-
ferences due to the Pll scores on the responses to all five egory of televisions. Perhaps the change in the wording
behavioral statements pertaining to involvement. These of the question led to the weak results for that product
overall significant results allow the interpretation of each category. Perhaps interest in how televisions work can
proposition separately. indicate interest in technology or interest in quality of
Search for product information. High involvement performance other than interest in the product per se.
consumers should be more interested in acquiring in- Alternative evaluation. One of the characteristics of
formation about the product than low involvement high involvement is the evaluation of competing alter-
consumers. For example, Engel and Blackwell (1982) natives. Since the highly involved consumer searches
defined involvement as the activation of extended for relevant information, the available alternatives are
problem solving behavior, and Bettman (1979) cited thought to be consciously compared before a selection

is made. To tap this dimension, subjects were asked the SENSITIVITY TO SITUATIONAL
extent to which they agreed with the statement "I have DIFFERENCES
compared product characteristics among brands of
." For all products, the high scorers had signif- The second content validity, the criterion validity,
icantly greater agreement with the statement than low and the construct validity sections have demonstrated
scorers. that the level of involvement with product categories
varies greatly over individuals. For any product cate-
Perception of brand differences. The next proposi-
gory, there seems to be individuals who have low in-
tion tested was that high involvement scorers would
volvement with the product and individuals who have
perceive greater differences among brands in the prod-
high involvement with the product. Additionally, the
uct class than low involvement scorers. This proposition
average level of involvement varies across the different
stems from writings of Robertson (1976), who suggests
products. For example, students rated bubble bath 69
that high involvement implies that beliefs about product
on the PII and rated automobiles 122 on the PII. This
attributes are strongly held, whereas low involvement
demonstrates that different products are perceived dif-
individuals do not hold strong beliefs about product
ferently by the same people. The scale is also proposed
attributes. Thus, the strength of the belief system to the
to be sensitive to different situations, a third factor that
attributes emphasizes the perception of differences
causes involvement, given the same people and the same
among brands on the attributes where beliefs are
strongly held. Subjects were asked to respond to the
Previous studies by Clarke and Belk (1978) and Belk
statement "I think there are a great deal of differences
(198 1) demonstrated that some purchase situations can
among brands of ." High scorers always per-
be more involving than others. They found that the
ceived greater differences (p < 0.01) among brands than
low scorers in the product class. purchase of some previously uninvolving products for
gifts can raise the level of involvement in the purchase
Brand preferences. People highly involved in a decision. To investigate the possibility of rating pur-
product class were hypothesized to have a most pre- chase situations on the scale, the Pll was administered
ferred brand in the product category. The preference over two purchase situations for wine to 41 members
of a particular brand stems from the perception of dif- of the clerical and administrative staff used in the pre-
ferences among brands. Since high involvement implies vious construct validity study.4 Each subject rated two
perceiving greater differences about product attributes, purchase situations: (1) the purchase of a bottle of wine
then the consumer should have a greater preference for everyday consumption, and (2) the purchase of a
based on that product differentiation. Again, over all bottle of wine for a special dinner party. The scale items
three products, high scorers showed a significantly (p were internally reliable for these purchase decisions.
< 0.01) greater agreement with the statement "I have Cronbach alphas were 0.98 and 0.97 respectively, and
a most preferred brand of " than low scorers. the item-to-total correlations were generally above 0.50.
In conclusion, the various measures of construct va- For this data collection, these situations were coun-
lidity used the correlation of two paper and pencil tests terbalanced across subjects. The mean scale score for
on the same subjects as evidence that the proposed scale the everyday consumption was 78 (s = 34), and for the
does tap the construct of involvement, as applied to special dinner party was 106 (s = 24). A related mea-
product categories. Although no one result is an excel- sures t-test was significant at t(40) = 5.42, p < 0.001;
lent test of the scale, each finding adds to the weight of therefore, the two purchase situations were rated dif-
evidence that the scale is an acceptable measure of in- ferently on the PIT.The analysis for differences between
volvement, as applied to product categories. the two purchase situations was also carried out as a
between-subjects design. Twenty-two subjects first rated
FACTOR ANALYSES OF THE PII the everyday consumption (X = 76) and 19 subjects
first rated the special dinner party (X = 98). The be-
An investigation of the dimensionality of the twenty- tween-subjects t-test for the first rating was also signif-
item scale was carried out for each product category icant at t(39) = 2.34, p < 0.05. These results show that
used in the scale development. The items were factor the PITis sensitive to different situations, if people and
analyzed using varimax rotation with squared multiple product remain constant. The PII may hold promise as
correlations in the diagonal for factor extraction. The a measure of involvement with purchase decisions, per-
general pattern of results showed one main factor and haps even applicable as a manipulation check for ex-
(usually) one minor or residual factor for every product periments that deal with manipulation of the situation
category. The major factor accounted for a range of as manipulation of involvement level (e.g., Park and
common variance from 65 percent for jeans to 100 per- Young 1983; Petty and Cacioppo 1979, 1981; Petty,
cet for instant coffee. Over all products, all items loaded Cacioppo, and Goldman 198 1).
positively on the first factor, which indicates that the
asumption of a simple linear combination of the scale 4The other 16 subjects did not receive these scales as part of their
items was not violated. questionnaire.

SUMMARY P1l than the everyday purchase situation. Further, some

preliminary research indicates that the Pll is an inter-
The purpose of the study was to develop a scale to nally reliable measure when applied to advertisements.
measure the construct of involvement. Hence, a se- However, more research needs to be carried out to verify
mantic differentialscale was developed to capturethe the stability and construct validity of the PII to adver-
concept of involvement for products.This PersonalIn- tisements and purchase decisions.
volvement Inventorywas developedover four data sets The Pll should have several benefits to the study of
of 268 undergraduatepsychologystudents;two datasets consumer behavior. It offers the potential of a valid
with49 MBAstudents;and two datasetswith 57 clerical instrument to replace the ad hoc and untested ap-
and administrativestaff members.The scale was dem- proaches that have previously been used in the field.
onstratedto have content validity by expertjudges at Since involvement is proposed to be a variable in the
two phases of the scale development: first, for the se- decision process, the Pll offers researchers a quickly
lection of items, and second, through classificationof administered tool, generalizable across product cate-
open-endedresponsesfrom subjects.The reliabilityor gories, that can be used as a covariate to other research
stability of the scale over time was checked over two questions. The ultimate test of the scale is whether or
subjectpopulationsfor an averagetest-retestcorrelation not the instrument can be used in empirical studies to
of 0.90. The criterion-relatedvalidity of the scale was test various aspects of involvement. I am conducting
checked by demonstratingagreementwith the orderof such research, and as others use the instrument and
variousproductsas found in previousstudies.The con- generalizable norms develop, its true validity will be
structvalidity-the test of the scale to theoreticalprop- determined.
ositions-was then carriedout. The scale was admin-
isteredto clerical and administrativestaffand covered
threedifferentproductcategoriesand severalstatements APPENDIX A
of behavior proposed to be representativeof involve-
ment. Overall threeproductcategoriestherewas a pos- Personal Involvement Inventory
itive relationshipbetweenthe scale scores and the sub-
jects' responses to the statements of theoretical prop- The following Personal Involvement Inventory is designed
ositions pertainingto involvement. to measure a person's involvement with products. To change
the instructions to measure involvement with advertisements
or purchase decisions, the words in the parentheses should
Limitations be changed accordingly. To measure involvement with ad-
Missingfrom this scale developmentare tests of con- vertisements, the words "various products they regularly pur-
vergentand discriminantvalidity. The tests of conver- chase or have purchased in the past" would be changed to
"the advertisements you have just seen (read)." To measure
gent validitywith anothermeasureof involvementwere involvement with purchase decisions, the words "various
not carriedout because at the time of this scale devel- purchase decisions people make" would be substituted.
opment no other general involvement measurein the The name of the object to be judged should be inserted at
literaturehad been tested for reliabilityand validity.5 the top of the scale page. Examples for three applications of
Testsof discriminantvalidityto the conceptof expertise the different contexts of the object are: (1) if the product was
or knowledge structure were carried out and are re- the object, then "red wine" would be judged; (2) if an ad was
ported in Zaichkowsky(1985). Pll scores were found the object, then "the ad for Gallo wine" would be judged;
to be unrelatedto expertisebut relatedto productuse. and (3) if the purchase decision was the object, then the "pur-
Furthertests of discriminant and convergent validity chase of a bottle of wine for a special dinner party" would be
need to be carriedout with respectto other constructs. judged. The reader is reminded that the construct validity of
the scale has only been supported for products.
In particular,the relationshipof Pll scoresto attitudes
should be furtherexamined, since severalitems on the
scale appearto be similar to a measureof attitudes. Instruction Page
Althoughthe usefulnessof the PITwas demonstrated
for products,the initial aim was to select items so that Instructions
the same scale might also be appliedto advertisements
or purchasedecisions. Some data were collected over The purpose of this study is to measure a person's involve-
differentpurchase decisions and showed that the Pl1 ment or interest in (various products they regularly purchase
was internallyreliablefor differentpurchasesituations or have purchased in the past). To take this measure, we need
for the same product. Additionally,the purchasesitu- you to judge various (products) against a series of descriptive
ations differedin their involvement scores,as would be scales according to how YOU perceive the product you will
expected;the special dinner partyscored higheron the be shown. Here is how you are to use these scales:

If you feel that the (product) that appears at the top of the
5There is currently some research translating this scale to French page is very closely related to one end of the scale, you should
for possible validation to an independently developed Likert scale. place your check mark as follows:

Unimportant x :. : : : Important FIGURE

Unimportant : : : : : x Important
If you feel that the (product) is quite closely related to one or 30 -
the other end of the scale (but not extremely), you should b
place your check mark as follows: 25

Appealing _ x: :: :_: Unappealing
or 15:
Appealing _: : x: Unappealing 10 .10~~~~~~~~~~~~~~~~~~~
. . . . . . .

If you feel that the (product) seems only slightly related (but
5.... ... ... ................................................
not really neutral) to one end of the scale, you should place
21 80 96 139
your check mark as follows: Theoretical Median
Low involvement mean High involvement
Uninterested x: : * : Interested 25% 69 90 25%
Sample mean
or (s 32) SCORES
Uninterested : x Interested
N= 751.
bo indicates end-point users 26 at low of 20; 12 at high of 140.
Important NOTE: Skewness = -0.626.

1. Be sure that you check every scale for every (product); do

not omit any.
2. Never put more than one check mark on a single scale. APPENDIX B
Make each item a separate and independent judgment. Work Distribution of Scores Over All Products
at fairly high speed through this questionnaire. Do not worry
or puzzle over individual items. It is your first impressions, To decide where cut off points for low and high in-
the immediate feelings about the items, that we want. On the volvement were on the scale, an overall distribution
other hand, please do not be careless, because we want your was tabulated and is presented in the Figure. This over-
true impressions. all distribution is based on data collected from 751 sub-
jects over 13 product categories shown in Table 3.
Any Questions? However, some of the subjects filled out the PITfor more
than one product, and thus are counted more than once.
Scale Page The overall PII mean for these products is 89.55,
(insert name of object to be judged) whereas the true theoretical mean is 80. This deviation
important : : : :_ : unimportant*
from the theoretical mean is most likely due to the
of no concern _: : : : of concern to me product-dependent nature of the distribution. No at-
irrelevant : relevant tempt was made to consciously select products that were
means a lot to me _: means nothingto me* thought a priori to be more or less involving to the sub-
useless_: : useful jects. It seems that the scale was developed, perhaps,
valuable : worthless* over products that were somehow more involving. Ad-
trivial_: : : :: fundamental dition of other products, such as nails or canned peas,
beneficial : not beneficial* that might not be involving to the subjects might push
matters to me _: : : _ doesn't matter* the mean toward the theoretical mean of 80.
uninterested interested
Deleted from the pictured distribution are the two
significant - insignificant*
vital : : : : superfluous*
end points 20 and 140 (these values are computed into
boring : : : interesting the mean scores). Twenty-six points were deleted at the
unexciting exciting low end of 20, and 12 points were deleted from the high
appealing unappealing* point of 140. These scores indicate that the rater only
mundane : : : : fascinating used the endpoints of one and seven to rate the product
essential : : nonessential* in question.
undesirable_: : : : _ : desirable The distribution derived from the data was used to
wanted : : unwanted* classify scorers into either low, medium, or high in-
not needed : needed volvement when comparison among groups of individ-
Indicates item is reverse scored.
uals was of interest. Low scorers were defined as those
Items on the left are scored (1) low involvement to (7) high involvement on the right. scoring in the first quartile of the distribution; they had
Totaling the 20 items gives a score from a low of 20 to a high of 140. scores ranging from 20 to 69. Medium scorers were de-

TABLE 3 Cohen, Joel B. (1983), "Involvement and You: 1000 Great

Ideas," in Advances in Consumer Research, Vol. 10, eds.
Richard Bagozzi and Alice Tybout, Ann Arbor, MI: As-
sociation for Consumer Research, 325-328.
Mean Pll and Marvin E. Goldberg (1970), "The Dissonance
Product score (s) Sample (N) Model in Post-Decision Product Evaluation," Journal
of Marketing Research, 7 (August), 315-321.
Instant coffee 66 (40) 57a
Bubble bath 69
Comrey, Andrew L. (1973), A First Course in Factor Analyses,
(38) 45b
Breakfast cereal 69 (29) 43C
New York: Academic Press.
Mouthwash 74 (28) 68 b Engel, James F., David Kollat, and Roger D. Blackwell (1978),
Red wine 82 (31) 45C Consumer Behavior, New York: Dryden Press.
Red wine 84 (31) 57a and Roger D. Blackwell (1982), Consumer Behavior,
Facial tissues 87 (26) 47b New York: Dryden Press.
Headache remedy 91 (25) 68b French, John W. and William B. Michael (1966), Standards
35mm camera 96 (26) 45C for Educational and Psychological Tests and Manuals,
Color TV 97 (30) 57a
Washington, D.C.: American Psychological Association.
Jeans 99 (21) 47b
Laundry detergent 103 (23) 57a Greenwald, Anthony G. and Clark Leavitt (1984). "Audience
Calculator 112 (16) 68 b Involvement in Advertising: Four Levels," Journal of
Automobile 122 (19) 47 b Consumer Research, 11 (June), 581-592.
Houston, Michael J. and Michael L. Rothschild (1978),
* Clerical and administrative staff. "Conceptual and Methodological Perspectives in In-
bUndergraduate psychology students. volvement," in Research Frontiers in Marketing: Dia-
c MBA students.
logues and Directions, ed. S. Jain, Chicago: American
NOTE: Theoretical mean = 80. Actual mean based on above products = 90 (s = 32).
Marketing Association, 184- 187.
Howard, John A. and Jagdish N. Sheth (1969), The Theory,
of Buyer Behavior, New York: John Wiley.
fined as those scoring in the middle 50 percent of the Hupfer, Nancy and David Gardner (1971), "Differential In-
volvement with Products and Issues: An Exploratory
distribution; they had scores ranging from 70 to 1 10.
Study," in Proceedings: Association for Consumer Re-
High scorers were defined as those scoring in the top search, ed. David M. Gardner, College Park, MD: As-
quartile of the distribution; they had scores ranging from sociation for Consumer Research, 262-269.
111 to 140. The distribution was also used for the de- Kassarjian, Harold H. (1978), "Presidential Address," in Ad-
velopment of a two-group classification when neces- vances in Consumer Research, Vol. 5, ed. H. Keith Hunt,
sary-i.e., those relatively high scores and those rela- Ann Arbor, MI: Association for Consumer Research,
tively low scores. The scale mean (89.55) was used as x1ii-xlv.
the break point for this grouping. By using this overall (1981), "Low Involvement: A Second Look," in Ad-
distribution as a guide to classification, some compar- vances In Consumer Research, Vol. 8, ed. Kent B. Mon-
ison is possible between low and high scoring subjects roe, Ann Arbor, MI: Association for Consumer Research,
on the PII. 31-34.
Krugman, Herbert E. (1962), "An Application of Learning
Theory to TV Copy Testing," Public Opinion Quarterly,
[ReceivedMarch 1984. Revised June 1985.] 26 (Winter), 626-634.
(1965), "The Impact of Television Advertising:
Learning Without Involvement," Public Opinion Quar-
REFERENCES terly, 29 (Fall), 349-356.
Belk, Russell W. (1981), "Effects of Gift-Giving Involvement (1967), "The Measurement of Advertising Involve-
on Gift Selection Strategies," in Advances in Consumer ment," Public Opinion Quarterly, 30 (Winter), 583-596.
Research, Vol. 9, ed. Andrew Mitchell, Ann Arbor, MI: (1977), "Memory Without Recall, Exposure Without
Association for Consumer Research, 408-41 1. Perception," Journal of Advertising Research, 17 (Au-
Bettman, James R. (1979), An Information Processing Theory gust), 7-12.
Of Consumer Choice, Reading, MA: Addison-Wesley. Lastovicka, John L. (1979), "Questioning the Concept of In-
Bloch, Peter H. and Marsha L. Richins (1983), "A Theoretical volvement Defined Product Classes," in Advances in
Model for the Study of Product Importance Perceptions," Consumer Research, Vol. 6, ed. William L. Wilkie, Ann
Joiurnal of Marketing, 47 (Summer), 69-81. Arbor, MI: Association for Consumer Research, 174-
Bowen, Lawrence and Steven H. Chaffee (1974), "Product 179.
Involvement and Pertinent Advertising Appeals," Jour- and David M. Gardner (1978a), "Components of In-
nalism Quarterl', 5 1 (Winter), 613-621. volvement," in Attitude Research Plays For High Stakes,
Clarke, Keith and Russell W. Belk (1978), "The Effects of eds. John C. Maloney and Bernard Silverman, Chicago:
Product Involvement and Task Definition On Antici- American Marketing Association, 53-73.
pated Consumer Effort," in Advances In Consumer Re- and David M. Gardner (1978b), "Low Involvement
search, Vol. 5, ed. H. Keith Hunt, Ann Arbor, MI: As- Versus High Involvement Cognitive Structures," in Ad-
sociation for Consumer Research, 313-318. vances in Consumer Research, Vol. 5, ed. H. Keith Hunt,

Ann Arbor, MI: Association For Consumer Research, Content and Context," in Advances in Consumer Re-
87-92. search, Vol. 8, ed. Kent B. Monroe, Ann Arbor, MI:
Loken, Barbara (1984), "Attitude Processing Strategies," Association for Consumer Research, 20-24.
Journal of Experimental Social Psychology, 20 (July), , John T. Cacioppo, and Rachel Goldman (1981),
272-296. "Personal Involvement as a Determinant of Argument-
Markin, Rom J. and Chem Narayana (1975), "Behavior Based Persuasion," Journal of Personality and Social
Control: Are Consumers Beyond Freedom and Dignity?" Psychology, 41 (November), 847-853.
in Advances in Consumer Research, Vol. 3, ed. Beverlee Ray, Michael L. (1973), "Marketing Communications and
B. Anderson, Ann Arbor, MI: Association for Consumer The Hierarchy-of-Effects," in New Models for Mass
Research, 222-228. Communication Research, ed. P. Clarke, Beverly Hills,
Mitchell, Andrew A. (1979), "Involvement: A Potentially CA: Sage Publications, 147-176.
Important Mediator of Consumer Behavior," in Ad- Robertson, Thomas S. (1976), "Low Commitment Consumer
vances in Consumer Research, Vol. 6, ed. William L. Behavior," Journal of Advertising Research, 16 (April),
Wilkie, Ann Arbor, MI: Association for Consumer Re- 19-24.
search, 191-196. Rothschild, Michael L. (1984), "Perspectives in Involvement:
and Jerry C. Olson (1981), "Are Product Attribute Current Problems and Future Directions," in Advances
Beliefs the Only Mediator of Advertising Effects on Brand in Consumer Research, Vol. 11, ed. Tom Kinnear, Ann
Attitude," Journal of Marketing Research, 18 (August), Arbor, MI: Association for Consumer Research, 216-
318-332. 217.
Nunnally, Jum C. (1978), Psychometric Theory, New York: Sheth, Jagdish N. and M. Venkatesan (1968), "Risk Reduction
McGraw-Hill. Process in Repetitive Consumer Behavior," Journal of
Olshavsky, Richard W. and Donald H. Granbois (1979), Marketing Research, 5 (August), 307-310.
"Consumer Decision Making-Fact or Fiction?" Journal Traylor, Mark B. (1981), "Product Involvement and Brand
of Consumer Research, 6 (September), 93- 100. Commitment," Journal of Advertising, 21 (December),
Osgood, Charles, George J. Suci, and Percy H. Tannenbaum 51-56.
(1957), The Measurement of Meaning, Urbana, IL: Uni- Tyebjee, Tyzoon T. (1979), "Response Time, Conflict, and
versity of Illinois Press. Involvement in Brand Choice," Journal of Consumer
Park, C. Whan and S. Mark Young (1983), "Types and Levels Research, 6 (December), 259-304.
of Involvement and Brand Attitude Formation," in Ad- Wright, Peter (1973), "Cognitive Processes Mediating Accep-
vances in Consumer Research, Vol. 10, eds. Richard Ba- tance of Advertising," Journal of Marketing Research,
gozzi and Alice Tybout, Ann Arbor, MI: Association for 10 (February), 53-62.
Consumer Research, 320-324. "Analyzing Media Effects on Advertising Response,"
Petty, Richard E. and John T. Cacioppo (1979), "Issue In- Public Opinion Quarterly, 38 (Summer), 192-205.
volvement Can Increase Or Decrease Message Relevant Zaichkowsky, Judith Lynne (1985), "Familiarity: Product
Cognitive Responses," Journal of Personality and Social Use, Involvement or Expertise?" in Advances in Con-
Psychology, 37 (October), 1915-1926. sumer Research, Vol. 12, eds. Elizabeth C. Hirshman
and John T. Cacioppo (1981), "Issue Involvement As and Morris B. Holbrook, Provo, UT: Association for
A Moderator of the Effects On Attitude of Advertising Consumer Research, 296-299.