You are on page 1of 40

DATA COLLECTION

This chapter is not intended to present and exhaust all data collection methods and statistical techniques in all their complexity but we selected the most used methods, algorithms, analysis techniques for data collection, summarizing and presentation, useful for taking business decisions.

2.1 Introduction Statistical observations have a major role in the well functioning and in the informational flow of the company. The orientation toward optimizing business communication goes together with the companys need to be more and more profitable in a tight competition business environment. If we understand Statistics as an operational tool that helps and facilitates the process of decisions making in a business organization, then we can also understand the place and role it has in the business communication system. Statistical reports, as a part of the external communication system of the company, are transmitted periodically to the local and national authorities. In the same time, internal reports issued on a regular basis are completing the statistical information and together, they set the foundations of the

Statistics for Business Administration

information system of the firms management. The purpose of transmitting statistical documents inside the company is different from the one of external communication. Implicitly these reports will have a different format, another type of presentation, another path of transmission, distribution, storage, and retrieving in the statistical analysis. No matter the type of the final statistical report to be transmitted, there is a unitary conception concerning the methods to obtain data which are about to be processed.

2.2 Measurement Levels Data collection procedure represent a necessarily condition for a statistical research to accomplish its goal and need to be done previously to summarizing and analysis data. Data collection process can address both categories of data: secondary and primary data. It is often called statistical observation. After defining the statistical population, the units, and variables, data collection principle is to reveal and record the values and categories characterizing all the population units, called statistical data. We are going to call statistical data the numerical values observed directly for a quantitative variable and the numbers associated with the categories of the qualitative variables if these categories can be ranked. For a pure qualitative variable the statistical data will be the number of times each category is occurring - the frequency of appearance for each category. It means that for each unit we need to measure the values of the variable characterizing the unit. Therefore, each variable can be measured according to a measurement level, also called scale. Conventionally there are 4 measurement levels determining four scale types: nominal, ordinal, interval, and ratio scale. The measurement level of a variable or the type of scale related to each type of variable comprised within a statistical investigation represents essential information for the researcher, which will establish the appropriate analysis technique for the available data. The scale type is determined by a set of

Data Collection

rules allowing assigning categories or values to the attributes or characteristics we are trying to measure about each member of the population. Nominal scale The scale called nominal is characterized by the use of symbols in order to represent possible categories of the variable. These symbols allow regrouping exclusively the population units into different classes without any idea of order or distance between classes that can be computed. The symbols purpose is only to label different categories or classes. The nominal scale role is mainly to identify rather than measure, evaluate quantities. We are using nominal scales in order to record the gender, the marital status, occupation, profession, colors and identification number of the unit (i.g. SKU: Stock Key Unit for production units). Ordinal scale The scale called ordinal is characterized by using symbols to express different categories in which the variable can occur, allowing grouping the units into classes as the nominal scale, and further more, allowing ranking these classes. The ordinal scale is establishing an order between classes and it allows establishing which class represents the maximum and the minimum value, which can be taken by the variable. At this point we still cannot measure the distance between two classes, which were ranked, increasingly or decreasingly. Examples of variables suitable to be measured on the ordinal scale are the social classes, the influence of the media on the public opinion, political opinions, behavioural patterns, attitudes, etc. Interval scale The numerical values assigned to the categories, the possibilities of occurrence for a variable on the interval scale allows grouping the units into classes, establishing an order between the classes, which are ranked, and also identifying the distances between classes. With the help of the interval scale distances between classes are compared and evaluated. Interval scale implies the existence of the measurement units. Variables as for instance the temperature, IQ index, the speed, are measured on the interval scale.

Statistics for Business Administration

Ratio scale The ratio scale implies in addition to the previous scales the existence of an origin point zero. Marking two points on the scale does not depend on the measurement unit. Among the variables treated on such a type of scale we can have the time, the length, and the revenue of a person. As noticed, there is a certain order between the categories of scales. They are presented successively from the simplest, the nominal scale, to the ordinal scale, interval scale ending with the ratio scale, the most complex one. The statistical analyze techniques depend on the type of scale suitable for each variable. Mardel and Nadeau suggest in Statistique en Gestion et en Economie, the following use of the measures and methods per each type of scale:
Statistical techniques applied according to the type of measurement scale Table 2-1 Scale type RATIO scale INTERVAL scale ORDINAL scale Statistical operations Measures of position Measures of dispersion Measures of association Contingency Coefficient Rank correlation coefficients, Significant tests. Chi-square sign test t test, Fisher test all the previous tests NOMINAL scale Mode Median, Quintiles Quintiles Arithmetic Mean Standard Deviation Correlation, regression Percentage of Variation all the previous methods Geometric Mean

Data Collection

2.3 Statistical Investigation Process The statistical observation or investigation is the general method of collecting the necessary data for managerial and market decision making. Generally, the notion of observation represents in Statistics the concrete and objective action used for obtaining data that is being recorded taking into account the requests of a corresponding statistical processing. In addition, observation can mean data recording with no contact with the unit. The statistical methods are aiming to eliminate, as far as possible, the difference between the economic reality in every business environment and the information necessary for the management system. Generally, the managers are asking for centralizing, typical, synthetic values instead of individual data. Business decisions count on a set of information (obtained from massive masses of statistical, financial, mathematical data) and last but not least, on the intuition of the decision maker. The information is obtained after a statistical investigation. We call statistical investigation any procedure of logical structure through which information is supplied to the decision-making system, and which relies on a complex analysis of primary and secondary data from the objective reality. A statistical investigation is organized in order to solve some major problems even less important ones, independent or connected to other investigations, organized periodically or occasionally. Statistical investigations are concrete statistical researches, which study the statistical population through exhaustive recordings/ records (when possible) and through collection of partial data. From this point of view, we will consider primary data data that is obtained through statistical observation and it is used for the first time. We will define secondary data as data from other sources (including banks of statistical data), data that are used for the 2nd time, at least, or for the nth time. This way, secondary data can be considered as being complementary to primary data gathered for the first time through the statistical data collection methods. Statistical investigations, being usually complex and significant operations, have to be organized in such a way, that they solve methodological and organizational problems.

Statistics for Business Administration

The statistical observations present a great variety. They can be classified according to several criteria, as, for example: a. After the organization manner we have permanent observations (the statistical reporting system or some periodical polls) and especially organized observations (inquiries, censuses, surveys) b. After the type of observation, they can be current, periodical, occasional, unique observations c. After the manifestation way the observations can be static, expressing the current status, measuring the volume and structure of phenomenon at a given time in a specific period and dynamic, recording the variation produced in the volume and structure of the collectivity necessary to establish the trend of development of the influence factors to the dynamics of the phenomena; the results represent the basis of making the predictions d. After the number of statistical units within the observation, it can be total (censuses and statistical reports) or partial (a selection or other statistical methods of collecting data). We will call an overall or total observation that particular statistical investigation which examines each unit of the statistical population. For example, in order to find out the opinion of the employees about the introduction of a new bonus and salary plan, all the employees are asked to answer a questionnaire especially created to collect pertinent information about this subject. The government of a country organizes also numerous overall observations called censuses. The census of the population is, probably, the most popular example in all countries of the world, being organized on a regular basis, approximately every 10 years. In Romania the last census was conducted in 2002. According to the Census of population and dwellings, information is obtained according to age, sex, occupation, structure of house-keepings, education, profession, number of rooms per family member and so on. Also, a census of the industry is organized in Great Britain, every five years, mining, constructions, agriculture and public utility services. Information obtained covered a large area of issues as: the optimal allocation of resources, the raw materials and finished goods stocks, production and maintenance costs and so on. An overall observation obviously presents the advantage of the data completeness and representativeness, and the disadvantage of consuming a

Data Collection

lot of time, labour and financial resources. Moreover, sometimes, the organization of a total research is impossible because of the destructive methods, during the quality control of products, or because of the impossibility to contact the entire population of reference, or simply because the members of the whole collectivity are not known (the potential customers of a certain product). In the economic and business environment data is often obtained after organizing a partial research, investigating selectively the statistical units forming a representative set from the general collectivity, called sample. Business success or failure depends on what the customers wish for, want, and buy. The answer stays for many business cases in the power of a research method called sampling. The information collected through sampling offer a good evaluation of the demand level, attitudes and opinions of the population from which the sample was taken from. The advantages of sampling are: - lower costs compared to the ones for the total research, in terms of time and resources, - possibility for the research to be conducted wherever is an objective impossibility to organize a total research. The main sample disadvantage is the possibility to obtain insignificant and insufficient data. Surely, the optimum way of obtaining information about a certain population is to organize a total research. When this is not possible (for example, within the quality study of products, these would be destroyed) we will organize a partial research. The statistical data collected during a sampling, inevitably, induces errors. The common steps for every type of statistical research - no matter the research type, the purpose for which it is organized respect the following procedure: 1. Establishing the objective of the research only if there is a management problem to be solved. Nobody is conducting a statistical research if there is no decision to be taken. 2. Defining and identifying the population to be studied according to the objective. 3. Establishing the set of characteristics or attributes according to the information we need to obtain. This set of characteristics is called the research program.

Statistics for Business Administration

4. Analyzing the data bases already existing about the population, in other words analyzing the secondary data sources. 5. If the secondary data is insufficient, the researcher has to decide to organize either a total research or a partial one, depending on the resources available such as time, money, and human resources. In some cases it is impossible to organize a census, like for instance for the production quality control in the case of a destructive control. If we decide to conduct a partial research, we need to decide what type of research we can conduct. 6. Organizing the data collection, which means deciding where, when and how to collect the data for each unit. How to collect data? We can choose to collect data individually, each unit on a separate fish or collectively using a common recording way like a list. 7. Data recording with the help of a data analysis program like SPSS, STATISTICA, SORITEC SAMPLER or MINITAB. 8. Data summarizing and presentation, also called primary data processing. The result of this step will be tables, series and graphs. 9. Data description analyzing using descriptive statistics and inferential measures. 10. Data conclusions presented into a research report. We may conclude that any statistical research has three main steps: data collection, data analysis and decision. All the data collection methods should have as premise the following characteristics: a. Objectivity meaning that it should be possible to measure and transmit both categories of data: numerical and qualitative data b. Variability of individual data through which the variable is recorded c. Repeatability the variables should have a certain repetitive character to be recorded consequently, in order to establish a central tendency into a data set d. Efficiency - the action of collecting data should keep account on the ratio between the involved resources, expenses, obtained results and the operability by the making decisions process. Data collected represents the raw material for obtaining the indicators allowing the statistical knowledge. That is why the statistical observation develops methodically some strict technical problems as: settling the methods and procedures of the observation, identifying the units of the

Data Collection

observation, of the multitude of statistical features subdue to the observation (the program of observation), of time and place of the observation, as well as the organizational details. There has to be mentioned that a clear distinction will be made in any statistical observation between the unit of observation and the reporting unit. For data recording we can use individual and/or collective formularies, being accompanied by instructions regarding the sampling program, the indicators needed and the recording manner. The statistical observation needs also to be defined in time through the settling of the period or of the moment of recording. The date around which data are being recorded, is considered to be the critical date of the recording meaning that the date written on the formularies will refer to this day. The place of observation is regularly the same as the place of development of the analyzed variable.

2.4 Secondary Sources of Data and their Use Secondary data can be classified according to the way they are collected or to their distribution sources. At the level of individual companies, secondary data are classified in external and internal data. Secondary sources of data comprise both external and internal statistical reports. Internal data is represented by operational reports and occasional ones that go together with any of the operations in the company: sales data, advertising reports, periodic inventory checking, and transportation costs. All of these are examples of internal secondary data for an organization together with its accounting and sale registers, personnel reports, etc. External data is the result of studies undertaken by specialized organizations and official organizations. This data comes from a wide range of external sources. Their number is enormous. Secondary data can be found in books, magazines, Governmental publications. Various secondary data types can be classified by the company that provided them, as follows: - Governmental sources, like publications of the central and local Statistics Institutions, reports from central Governmental Institutions or information from the local power, organizations or international Institutions.

Statistics for Business Administration

- Commercial sources, information coming from the Chamber of Commerce or other consultancy organisms - Sources of district Chambers - Academic reports from other institutions. Among them, Governmental sources, reports, and publications of the National Institute for Statistics and of the Register of Commerce are the most credible sources.

2.5 Methods of Collecting Primary Data. Interviewing Methods Collecting data represents the method of obtaining information from the people who are being interviewed in a statistical survey. According to the observation method, the identifiers of the subjects, the available resources and the degree of accuracy, we can use different methods to collect data. In the following paragraphs some methods of collecting data are presented together with the advantages and disadvantages for each one. Once the researcher has decided on the sample he or she must decide on the method to gather data, taking into account each method advantages and disadvantages. 2.5.1 Personal Interviews An interview is called personal when the interviewer asks the questions face-to-face with the interviewee. Personal interviews can take place at the respondent home, at a shopping mall, on the street, outside a movie theater or polling place, in the vehicle, and so on. Advantages: Maximum information is obtained The ability to record the respondent reaction to the interview The ability to let the interviewee see, feel and/or taste a product or pictures of the product High response rate The ability to find easier the target population. For example, you can find people who have read a book more easily outside a book shop than by calling randomly phone numbers. Longer interviews are easier tolerated, particularly for in-home interviews that have been arranged in advance. People may be willing to talk longer

Data Collection

face-to-face than to someone on the phone. This is due to the physical contact with the survey operator. Disadvantages Personal interviews usually cost more per interview than other methods. This is particularly true of in-home interviews, where travel time is a major factor. As average a survey operator full time employed is paid with around 800 USD/week Time! Sometimes it takes to long to travel to the respondent home Influence of the operator on the respondent Bias: for instance when interviewing into a mall we should take into account that each mall has its own characteristics. It draws its clientele from a specific geographic area surrounding it, and its shop profile also influences the type of client. These characteristics may differ from the target population and create a non-representative sample. 2.5.2 Telephone Interviews Surveying by telephone is the most popular interviewing method particularly in the USA. This is made possible by nearly universal coverage (96% of homes have a telephone). Advantages Low cost for local surveys High response rate, that can be improved by repeated calls People can usually be contacted faster by phone than by other means. If the interviewers are using computer-assisted telephone interviewing, the results can be available minutes after completing the last interview. High speed of data recording, coding, tabulation and interpretation A low time needed for this type of research You can dial random telephone numbers when you do not have the actual telephone numbers of potential respondents. In a phone interview software makes complex questionnaires practical by offering many logic possible options. Computers can control the interview better than humans. It can automatically skip questions, perform calculations, and modify questions based on the answers to earlier questions. It can check the logical consistency of answers and can present questions or answers in a random order

Statistics for Business Administration

Disadvantages Not everybody has a fix phone! Not everybody is listed in the phone book The list of mobile phones is not available Hugh costs for national or international surveys The relative small duration of the interview. You can ask a small number of questions. These should be simple and easy to understand Many companies have given the legitimate research a bad name by claiming to be doing a survey when they do a commercial call Consequently, many people are reluctant to answer phone interviews and use their answering machines to screen calls. The growing number of working women often means that no one is home during the day. This limits calling time to a break between 5-9 p.m. when you might interrupt dinner or a favorite film. So it is recommended to call during commercial breaks. You cannot show pictures or sample products by phone. 2.5.3 Mail Surveys Advantages Mail surveys are among the least expensive Mail can reach everybody anywhere with a few exceptions This is the only kind of survey you can do if you have the names and addresses of the target population, but not their telephone numbers. The questionnaire can include pictures No influence from the survey operator. This allows the respondent to answer at their leisure, rather than at the often inconvenient moment they are contacted for a phone or personal interview. Disadvantages It takes longer than other surveys to gather data by mail. You will need to wait several weeks after mailing out questionnaires before you can be sure that you have gotten most of the responses. The questionnaire has to be as short as possible, cannot comprise as many variables as a personal interview Low rate of response In populations of lower educational and literacy levels, response rates to mail surveys are often too small to be useful. This eliminates a lot of

Data Collection

low-educated people in many areas. Even in well-educated populations, response rates vary from as low as 3% up to 90%. As a rule of thumb, the best response levels are achieved from highly-educated people and people with a particular interest in the subject (which, depending on your target population, could lead to a biased sample). One way of improving response rates to mail surveys is to mail a postcard telling your sample to watch for a questionnaire in the next week or two. Another is to follow up a questionnaire mailing after a couple of weeks with a card asking people to return the questionnaire. The downside is that this doubles or triples your mailing cost. If you have purchased a mailing list from a supplier you may also have to pay a second (and third) use fee - you often cannot buy the list once and re-use it. Another way to increase responses to mail surveys is to use an incentive. One possibility is to send a EURO bill (or more) along with the survey (or offer to donate the dollar to a charity specified by the respondent). If you do so, be sure to say that the Euro is a way of saying "thanks," rather than payment for their time. Many people will consider their time worth more than a Euro. Another possibility is to include the people who return completed surveys in a drawing for a prize. A third is to offer a copy of the (non-confidential) result highlights to those who complete the questionnaire. Any of these techniques will increase the response rates. 2.5.4 Computer Direct Interviews These are interviews in which the interviewees enter their own answers directly into a computer. They can be used at supermarkets, trade shows, offices, and so on. Advantages The virtual elimination of data entry and editing costs. Overall low costs High speed and low time for data gathering You will get more accurate answers to sensitive questions. Recent studies of potential blood donors have shown respondents were more likely to reveal diseases related risk factors to a computer screen than to either human interviewers or paper questionnaires.

Statistics for Business Administration

Computer-aided surveys among drug users get better results than personal interviews. Employees are also more often willing to give more honest answers to a computer than to a person or paper questionnaire. Another advantage is the elimination of interviewer bias. Different interviewers can ask questions in different ways, leading to different results. The computer asks the questions the same way every time. Ensuring skip patterns are accurately followed. These automatic skips are more accurate than relying on an interviewer reading a paper questionnaire. Response rates are usually higher. Computer-aided interviewing is still novel enough that some people will answer a computer interview when they would not have completed another kind of interview. Disadvantages The interviewees must have access to a computer or one must be provided for them Impossibility to have long questionnaires allowing to record a lot of variables Low response rate in populations of lower educational and literacy levels. This method may grow in importance as computer use increases. 2.5.5 E-mail Surveys E-mail surveys are both very economical and very fast. More people have email than have full Internet access. This makes e-mail a choice together with the Web page survey, particularly for young populations. On the other hand, e-mail surveys are limited to simple questionnaires, whereas Web page surveys can include complex questionnaires. Advantages Speed. An e-mail questionnaire can gather several thousand responses within a day or two. No cost involved once the set up has been completed. Possibility to attach pictures and sound or motion files. It is more fashionable than ordinary old fashion mail surveys. Disadvantages Difficult access to a list of e-mail addresses Multiple responses. Many programs have no check to eliminate people responding multiple times to bias the results. The programs should only

Data Collection

accept one reply from each address sent the questionnaire. It eliminates duplicate and pass along questionnaires and checks to ensure that respondents have not ignored instructions (e.g., giving many answers to a question requesting only one). Many people dislike unsolicited e-mail even more than unsolicited regular mail. You may want to send e-mail questionnaires only to people who expect to get e-mail from you. Virus threat fear! Cannot make inferences about the population Many e-mail programs are limited to plain ASCII text questionnaires and cannot show pictures. Many average citizens still do not possess e-mail facilities. Therefore, e-mail surveys do not reflect the population as a whole. At this stage they are probably best used in a corporate environment where e-mail is much more common or when most members of the target population are known to have e-mail.

2.5.6 Web- page Surveys Web surveys are rapidly gaining popularity. They have major speed, cost, and flexibility advantages, but also significant sampling limitations. These limitations make software selection especially important and restrict the groups you can study using this technique. Advantages Extremely fast. A questionnaire posted on a popular Web site can gather several thousand responses within a few hours. Many people who will respond to an e-mail invitation to take a Web survey will do so the first day, and most will do so within a few days. After the set up has been completed there is no other cost. Large samples do not cost more than smaller ones You can show pictures and motion sound films Web page questionnaires can use complex question skipping logic, randomizations and other features not possible with paper questionnaires or most e-mail surveys. Web page questionnaires can use colors, fonts and other formatting options not possible in most e-mail surveys.

Statistics for Business Administration

On average, people give longer answers to open-ended questions on Web page questionnaires than they do on other kinds of self-administered surveys. Some Web survey software can combine the survey answers with pre-existing information you have about individuals taking a survey. Disadvantages Current use of the Internet is far from universal. Internet surveys do not reflect the population as a whole. People can easily quit in the middle of a questionnaire. They are not as likely to complete a long questionnaire on the Web as they would be if talking with a good interviewer. If your survey is on a web page, you often have no control over who replies - anyone from Antarctica to Australia, cruising that web page may answer. Depending on your software, there is often no control over people responding multiple times to bias the results. At this stage we recommend using the Internet for surveys mainly when your target population consists entirely of Internet users. Business-tobusiness research and employee attitude surveys can often meet this requirement. Surveys of the general population usually will not. Another reason to use a Web page survey is when you want to show video or both sound and graphics. A Web page survey may be the only practical way to have many people view and react to a video. In any case, be sure your survey software prevents people from completing more than one questionnaire. You may also want to restrict access by requiring a password or by putting the survey on a page that can only be accessed directly. 2.5.7 Scanning Questionnaires Scanning questionnaires is a method of data collection that can be used with paper questionnaires that have been administered in face-to-face interviews; mail surveys or surveys completed by a survey operator over the telephone. Advantages Scanning can be the fastest method of data entry for paper questionnaires. Scanning is more accurate than a person in reading a properly completed questionnaire.

Data Collection

Disadvantages Cannot be used for open ended questions Requires pre-coding operation Scanning is less accurate than a person in reading a poorly marked questionnaire. Requires investment in additional hardware 2.5.8 Summary of Survey Methods Your choice of survey method will depend on several factors. These include:
Speed

Email and Web page surveys are the fastest methods, followed by telephone interviewing. Mail surveys are the slowest.

Cost

Personal interviews are the most expensive followed by telephone and then mail. Email and Web page surveys are the less expensive for large samples as compared to other surveys.

Internet usage

Web page and Email surveys offer significant advantages, but you may not be able to generalize their results to the population as a whole.

Literacy levels

Ignorant and less-educated people rarely respond to mail surveys.

Sensitive

People are more likely to answer sensitive questions when

questions interviewed directly by a computer in one form or another. Video, sound, graphics

A need to get reactions to video, music or a picture limits your options. You can run a video on a Web page, in a computer-direct interview, or in person. You can play music when using these methods or over a telephone. You can show pictures in those first methods and in a mail survey.

Statistics for Business Administration

The advantages and disadvantages of each technique are synthesized by George Kress in Marketing Research, Prentice Hall, NY, 1994 (page 90) in Table 2-2.
Advantages and disadvantages of data collection methods Table 2-2
Types/methods of collecting data Criteria Probable rate of response Control of the interviewers influence Utilization questionnaires Costs Speed of receiving answers Suitable sample size Sample representativeness good low good acceptable good excellent acceptable good acceptable acceptable low excellent Good Excellent Good acceptable of complicated low good excellent Low Mail acceptable excellent Telephone acceptable acceptable Personal acceptable low Computer acceptable Acceptable

2.6 Questionnaire Design 2.6.1 General Considerations The first rule is to design the questionnaire to fit the medium. Phone interviews cannot show pictures. Survey-by-mail respondents cannot ask, What exactly do you mean by that question? if they do not understand a question. Intimate, personal questions are sometimes best handled by mail or computer, where anonymity is most assured. It is not usually recommended that combine methods. A mail survey will often not give the same answers as the same survey done by phone or in person. If you used one method in the past and need to compare results, stick to that method unless there is a compelling reason to change. One of the most important rules is: MISS maintain it as short and less sophisticated as possible. If you present a questionnaire with 100 questions most potential respondents will give up before even starting. Ask yourself what you will do with the information from each question. If you cannot give yourself a satisfactory answer, leave it out. Avoid the temptation to add a few more questions just because you are doing a questionnaire anyway.

Data Collection

If necessary, place your questions into three groups: must know, useful to know and nice to know. Discard the last group, unless the previous two groups are very short. Always start with an introduction or Welcoming Message. In the case of mail questionnaires, this message can be in a cover letter or on the questionnaire form itself. If you are sending e-mails that ask people to take a Web page survey, put your main introduction or welcome message in the e-mail. When practical, state who you are and why you want the information in the survey. A good introduction or welcome message will encourage people to complete your questionnaire. Allow a Don't Know or Not Applicable choice of response to all questions, except to those in which you are certain that all respondents will have a clear answer. In most cases, these are wasted answers as far as the researcher is concerned, but are necessary alternatives to avoid frustrated respondents. Sometimes Don't Know or Not Applicable will really represent some respondents' most honest answers to some of the questions. Respondents, who feel they are being forced into giving an answer they do not want to give, often do not complete the questionnaire. For example, many people will abandon a questionnaire that asks them to specify their income, without offering a "decline to state" choice. For the same reason, include Other or None whenever either of these are a logically possible answer. When the answer choices are a list of possible opinions, preferences or behaviors you should usually allow these answers. On paper, computer direct and Internet surveys these four choices should appear as appropriate. You may want to combine two or more of them into one choice, if you have no interest in distinguishing between them. You will rarely want to include Don't Know Not Applicable, Other or None categories in a list of choices being read over the telephone or in person, but you should allow the interviewer the ability to accept them when given by respondents.

Statistics for Business Administration

2.6.2 Steps for the Questionnaire Development For a structured research, the questions are precisely asked and arranged in an already settled manner. In order to develop the questionnaire, the following steps proposed by Kress G., for a marketing research and adapted for any statistical investigation, have to be followed: Step 1: Determining the specific necessary information and the way of utilization An often mistake in the poll practice is the attitude: Have I wanted to ask something more? Once the questionnaire is finished and a number of interviews are taken, it is much too late to include other questions (Henry Skinner Marketing). That is why before developing the questionnaire, the researcher has to identify the whole of specific data necessary to the reach of the objective of the study, and to decide how the information resulted from the processing of data will be used. For analysis the researcher can use methods as the regression or factorial analysis, techniques that will be presented later on, in a separate chapter of Regression and Correlation. Very often, the researchers make mistakes by not taking care of these problems, knowing that there is the risk that the collected data does not match with the analysis technique wanted for obtaining a certain type of results. Step 2: Selection of the interviewing procedure What study method is most appropriate for gathering of data? The decision is connected to three key elements: the type of searched data, the type of persons, which will be interviewed, and their position within the firms that are part of the sample, if the sample is formed by organizations. Referring to the data included in the statistical observation program, there has to be mentioned if the information can be obtained through a formal, structured questionnaire, or has to be obtained through a poll process that is unstructured, undefined, or disguised? Referring to the statistical units included in the sample, there has to be mentioned the location of the subjects, the area of the study, the position of the persons that are about to be interviewed their accessibility and willingness to be interviewed. About the conditions of time and cost, what the researcher wishes to accomplish can be very much different from the researchers aim, depending on his financial possibilities.

Data Collection

Step 3: The selection of the questions content When we are examining every question that we want to put in the questionnaire, it is good to ask ourselves if the people answering have access to the necessary information in order to answer it correctly. It was noticed that some persons, even if they are aware of their inability to answer, do not give up on responding and even answer with explanations. They will express opinions about problems that havent crossed their minds or they understand very little of. These difficulties are not referring only to opinion questions or organized polls on samples of consumers but also to samples formed by organizations. When asking an employee for factual information, it is necessary for him to be in the right position, which permits that his answer is competent, responsible and for him to have the ability to transmit his answers correctly. In conclusion, the researcher should: address only to those persons which can give correct answers, refer within the question just at past events only if he can expect the other person to remember them and to have access to the support documentation and ask for opinions only if these have the appropriate responsibility/competence inside the company included in the sample. Willingness is not a guaranty of good answers. For example, in order to avoid this situation, within a poll aiming to evaluate the training policy, the interview was taken with the manager of human resources assumed to be the most appropriate employee to present the policy of personnel developing and training, and the intentions of the firm in this direction/field. An interesting observation is made by Moser: people may have the tendency to decrease the age and consume of alcohol, but on the other hand, they tend to exaggerate their preparation, readings, going to the church. By extension, we can consider that the mangers of human resources can tend to decrease their losses, large flows of personnel and they will exaggerate with the number of trained persons, the number of courses organized. Authors agree that two main categories of questions can be found in a questionnaire. After the content we will classify the questions into factual and opinion questions which can be enclosed depending on the method chosen for the data collection.

Statistics for Business Administration

a) Factual questions regard facts, realizations. There are such questions about the motivation or about the possibility of information, but the methodological aspects will be emphasized if we examine different sorts of factual questions and the opinion ones. A major difference between them consists of the degree of freedom, which is allowed to the interview operator to explore, explicitly, to formulate the questions as varied as possible. At the factual questions (if the person questioned is in direct contact with the facts, with the products he knows or with the events) the interview operators can take any reasonable measures to assure understanding and correct registering of the questions, and respectively the answers. In the case of opinion questions, such freedom would be too risky and that is why it is not allowed. It is known that the changes in the formulation, the succession, or the putting in evidence of the questions, the intonation, can have a real effect changing completely the answers. Some researchers do not allow the interview operators more freedom for factual questions, considering that the reality depends very much on a precise definition that this would be too risky (Kress Marketing Research, Prentice Hall, NY, 1994). In the poll made, the majority of the questions in the questionnaire refer to facts, to the behavior and to the policy of the firm regarding the development of the personnel through training and continuous forming. Moser uses the word fact in a general meaning. The question may be called factual although the answers given by the managers can be a mix of facts, vague hopes and of the wish to give an answer, which they consider to be at the operators expectations. This does not have such a big importance. The adjective factual refers here to the type of statistical data requested, not at the preciseness of the answer, so that the name factual of a question does not imply the preciseness of the answers. For example, the factual questions are those questions regarding the dominant form of property, the source of the capital, the activity area of the organization, the position of the employee, inclusion of the expenses for training courses within the company budget, the categories of employees which were included in professional preparation, the system of employee modeling, the work and life conditions of the employees. Some factual questions are called classification or identification questions and are designed especially to obtaining information through which the analysis

Data Collection

may distinguish the main groups of economical agents, by types and categories, out of those included in the poll. The classification questions have to be attentively defined and explained through separate instructions or not, by the rest of the questions usually positioned at the end of the interview for avoiding the agglomeration of personal questions in the first minutes. At any moment, the factual questions need a special introduction. That is because the person required to answer the questions regarding the training system at the firm he works at, can be amazed by the fact that he has to give details about the budget of revenues and expenses, or his preoccupations, his income or his age. For this, it is recommended for example a means of introduction put in the classification section as the following: Considering answering is anonymous, we ask if you can answer some necessary questions for the classification or Because of the fact that the training activities vary from a company to another, after the area of action, the general work conditions, the adherent market share and not the least after the size and power of the firm, we would like to find out some of the characteristics of your company, assuring you of the confidentiality of your answers The main difficulties raised by the factual questions are the making sure of the operators understanding of their exact purpose and the exact transmission to the interlocutor of the facts that they would like to know. This difficulty can be overcome through a structured study, for the development of which the operator is asked to know by heart the questions and to use during the interview, just the auxiliary materials lists, scales, tickets. With all the difficulties due to the cheating definitions, the chances that the operator or the person interviewed to misunderstand the question or not to understand it at all or - in extreme cases the answer to be influenced by the formulation of the questions, are less probable than in the case of opinion questions. b) In the field of opinion questions, the problems are much more complex and implicitly more difficult to avoid and to be corrected. Strictly connected to this problem, is the problem of the intensity of the opinion. For about any subject given, some persons have intense feelings while others are indifferent or are not interested, some have stable and coherent opinions and others have an extremely mobile attitude. That is why, in order to appreciate correctly the intensity of the attitude and opinion towards for example the use of computers in different areas, and to obtain a rightful

Statistics for Business Administration

segmentation, by types of managerial behavior and personality by company type, there can be used different scaling methods, which help ranking the opinions. Such an example can be from -2 (totally against the affirmation) to +2 (completely agree with the affirmation). We would like also to underline that the answers to opinion questions are much more sensitive to the changes in formulation, succession or putting into evidence than the factual questions (Moser, International Journal of Opinion and Attitude Research). Some standards have to be respected, in creating the questionnaire, for every question included. Is the question in general well asked and if so is it positioned appropriately within the questionnaire? Is the question associated with the objective of the study? Could the information be the same or quantitatively different if it is obtained through another method? It is important that every question offer the maximum quantity of information by consuming the minimum quantity of time and space. The researcher should not emphasize the questions needing a high level of concentration and attention, as for example: We present to you a series of courses for professional development and personnel training. Please identify what areas are of interest to your organization, is a question which is not recommended to be placed at the beginning of the questionnaire Placing it at the end of the questionnaire we should make the distinction between the initiated persons that know about the subject and those that are not initiated. In addition, explanations need to be done to the subject of what it is all about. Step 4: The selection of the general types of questions, after their way of presentation In general, the questions are grouped regarding their form: a) open questions; b) questions with pre-coded answers c) multi-choice questions d) dichotomic or binary questions The researcher has to decide which type will be used for each factual and opinion question.

Data Collection

We can have many categories: a) Open questions. The questioned subject may answer them by using their own words and can express any idea suggested by the question. The requested answer can be, depending on the question, short (Which is in your opinion the major disadvantage of the advertising of training courses?) or extended answers (What do you think are advantages of the advertising made for the training courses?). Because the opened questions tend to suggest alternative answers, they tend to have less influence on the actual answer then the multiple answer questions do or the dichotomy ones. A weakness of this type of question is that the person questioned can misinterpret the searched information. Also there may appear difficulties in the subsequent codification of the answers, which is done manually for the opened questions, this being done through the procedure of stripping of the questionnaires and content analysis. Lots of answers are structured in such a manner that obtaining correlations is difficult. On the other hand, opened questions give importance to the answers of the persons with a certain level of education, usually with an over the average level, because this group is more articulate in their thinking and that is why they can communicate their ideas more efficiently. b) open questions with pre-coded answers. The subject can answer at these questions with their own words, the answers being enclosed in a variant codified before the development of the questions by the operator. The number of the variants for coding the answers is limited at maximum 1012 variants as recommended by George Kress in Marketing Research, Prentice Hall, NY, 1994. For example: Which are the advantages of the training policy for your company? - it helps maintaining the professional level of the employees (1) - it helps improving the professional level of the employees (2) - it determines the increase in efficiency of the employees (3) - it determine the increase in efficiency in general of the firm (4) - it provides new information to the employees (5) - it provides information about the present (6) - it increases the socializing degree of the employees (7) - it makes the working environment more colorful and alive (8) - something else, mention what it is __________ (9)

Statistics for Business Administration

b) Multiple-choice questions. In this case the subjects are offered a number of specific alternatives from which they may chose one or more answers. If it is possible, the listed alternatives should not interfere with one another. Nevertheless, the option Something else, please mention what it is: ______ or I dont know. This type of question gives the possibility of obtaining some rapid and efficient answers. That is why these are preferred by the persons questioned (Kress, Marketing Research, Prentice Hall, NY, 1994) and by the specialized researchers in the analysis of data. The elaboration of the processing models and the classification of the recorded data are very much simplified and the alternative answers guide the subject towards the expected answer. An unwanted aspect of this type of question is that the answers may be polarized because of the order of the answers. If one answer is requested, this type of question does not distort always the results, as for example: What does the training of the personnel mean to you? - a hobby (1), - a work instrument (2), - a secondary interest (3), - a main interest (4), - the future (5), - a purpose in life (6), - a purposeless activity (7). Pay attention! Only one answer has to be checked. The disadvantage of admitting multiple-choice pre-coded answers was proved to be the polarization of the answers towards the first options. The potential polarization caused by the position of the possible answers in the questionnaire, can be avoided through alternating the answers in the answer list, in the different sets of questionnaires. If the possible answers have a quantitative form, the nature of the polarization changes, the undecided managers tending to select the middle numbers rather then the extremes. It is difficult to eliminate this deficiency through rotating the numbers at different questions because this may lead to confusing the person in question. A third type of polarization includes the situations in which the letters a, b, c, d are used, in which case, it was statistically proven that the undecided persons choose the letter a in the great majority (this being associated

Data Collection

mentally with the highest performance) or the answers codified with a vowel, against the consonants. It is recommended that the letter to be attributed to the last option or the least probable one. Another method of removal of polarization, can be the use of a combination of letters as h, i, m (Kenneth Coney, - The special case of Letter of preference). Examples for different types of questions:

e) Binary questions. A binary or dichotomic question is an extreme form of a multiple-choice question, where there are only two possibilities of choice: Yes/No, Credit/Cash. Almost all the time there has to be included in the questionnaire a third option of answer: I dont know or Something else. The questionnaire starting with a dichotomy question wish to identify from the beginning the appropriate persons to take part at the poll. Step 5: Choice of the type of scales To approach the analysis and the evaluation of the attitudes, the opinions so to analyze the qualitative data, separately and most important connected to other qualitative and quantitative variables, for evaluating the factors that are being influenced and which influence such variables, the poll researchers use the scaling analysis (or hierarchic investigation) as method of statistical study.

Statistics for Business Administration

The scaling analysis allows spacing out, hierarchies of attitudes and their measurement. Through spacing out we will understand the qualitative understanding of the attitudes according to a hierarchy of the intensities of the opinions. Through measuring we are trying to evaluate the qualitative measurement of attitudes. One of the main problems appearing in the scaling analysis is the construction of the appropriate scale for the variable that is studied. Once the scale was chosen, it appears the problem of just positioning of the attitudes in conformity with that scale, an operation that is developed during the filling in of the questionnaire. Rating Scales and Agreement Scales are two common types of questions that some researchers treat as multiple choice questions and others treat as numeric open end questions. Examples of these kind of questions are:

Data Collection

Step 6: Decisions upon the drafting of the questions Some aspects of question formulation are very important in the marketing studies based on polls, more precisely: a) questions with a general aspect. Using a general question when a specific answer is requested represents a common error. A question like Are you satisfied with the personnel policy at your firm? is not satisfactory if the interest of the researcher looks for the friendship degree inside the human resources department. b) simple language. If the study is done on a sample of firms, the questions have to be formulated clearly, with no ambiguities, so that the requested data is no secret for the company. Simple does not mean simplistic. If the data are secret there confidentiality has to be guaranteed. If the subjects are a particular group, as engineers, doctors, professors, managers, the language can be technical and elaborated. The meaning of each question has to be clear. How often do you use in your activity the services of a specialized firm? can mean how many hours each day, or week, or month. Sometimes it is better to use two or three questions, in stand of one with multiple-answers. Harder to avoid are the words used every day by the experts, which are not part of the regular conversation of human resources managers. In order to assure that the questions are understood the reactions of the potential and typical questioned persons in a pilot study. Payne mentions a series of words to be replaced with simpler alternatives: to communicate = to inform to give assistance = to help to initiate = to begin to be materialized = to be produced, to have place major = essential, main, important The fundamental principle of formulating is that the questions have to contain the simplest words to communicate the exact meaning. The expression has to be as simple and clear as possible, with no place left for

Statistics for Business Administration

interpretation. The closer the questions are to the regular conversation, the more fluent goes the interview. Regarding the technical and scientifically language, the researcher has to be sure that all organized groups understand through one word the same meaning. A simple contradictory case is the expression human resources, which means for a part of the population the equivalent of the personnel department in an organization while for the other part it represents the labour resources of a country. c) ambiguity. Ambiguous questions need to be avoided. If any ambiguous word slips, then different persons will understand the question in different manners so they will answer accordingly. d) vague sense words. Vague questions provide vague answers. If the question is: What kind of technical support do you use at your office?, some employees will answer almost for sure pretty old, others will mention American, the managers will answer almost surely the best on the market because the name of the firm is at stake and others will just write the title and the publishing house of the manual. Vague answers will be obtained also when beginning the question with Why?. f) partial questions. Are those that through their content, structure of formulation suggest to the person questioned a certain answer. For example, at the question Do you agree/ dont you think that would suggest a negative answer and the question Shouldnt we will suggest a positive answer. Also, a question which suggests only some possible answers can direct the speakers, and that is why at the multiple- choice questions there always has to be attached the option Something else, please mention what or I dont know. g) presumptuous questions. Generally, the questions should not assume anything about the subject. They dont have to involve the fact that they necessarily have certain information or a certain opinion about the questioned subject.

Data Collection

Step 7: Decisions about the sequence of the questions The questions have to come one after the other in a logical sense, which assumes that the researcher has to respect the logics of the operator as well as the logics of the person interviewed. Both of them have to be as close as possible to the initial logics of the researcher, of what he initially meant. Any distortion would lead to collecting divergent information, and would mean a waste of resources or affecting the level of concentration due to the ambiguity. Usually, at opened questions it is easier to answer if they are placed right at the beginning of the questionnaire and not inside of it. Both the operator and the subject have to be capable to pass from one question to another easily. This is especially important when it is necessary to avoid some questions because of previous answers. After introducing a theme it is massively exploited the use of successive questions and not rare ones. Placing the questions on the same theme in a concentrated group concentrates the attention of the subject on a sole problem in time, this way the information being obtained more precise. This sort of arrangement, by sequels makes the subjects more comfortable because they see a logical way of approaching the questionnaire. Step 8: Maximizing the utility and attractiveness of the questionnaire The questionnaires have to be visually attractive, pretty short in order not to bother the subject. To increase the utility of the questionnaire the questions regarding the type of service or product are placed in the first part of the questionnaire. A pretty big proportion proves to be interested in speaking about their own tastes and life style, this is why it is recommended for increasing the attractiveness of the questionnaire the use of amusing scales, colorful pictures and all these before the section of the classification questions, which usually arises suspicions to the persons questioned. People are generally open to transmit information regarding the firm where they work especially if they promote the image of it but are reserved if they are asked for market segmentation information and for their own demographic characteristics. If we want to obtain demographical data, the

Statistics for Business Administration

questions regarding these issues are placed at the beginning of the questionnaire for the companies and at the end for individuals. It is also depending on the selection method. For quota sample this part is placed at the beginning of the questionnaire and for random sampling it is located at the end. Quota and random sampling are explained in Chapter 6. The structure of the questionnaire has to be thought in such a manner making easier the codification of the answers, as well as their preliminary processing, the obtaining of total indicators. Personal, difficult questions are usually placed at the end of the questionnaire. It is recommended that the operators learn by heart the questions so they are not forced to read during the interview, this will look much more like a free discussion. The only papers with which there will be work with will be the auxiliary materials. Step 9: The organization of a pretest the pilot study for testing the efficiency of the questionnaire. Before using the questionnaire as a tool for collecting primary data any questionnaire needs to be tested. The purpose of the organization of a pilot study has to be similar to the one of the organization of the final one. The sample for the pilot study can have a small volume although the sample projected for the poll is normal. Ideally the survey should be tested on the same kinds of people you will include in the main study. If that is not possible, at least have a few people, other than the question writer, in order for them to try the questionnaire. This kind of test run can reveal unanticipated problems with question wording, instructions to skip questions, etc. It can help to see if the interviewees do understand your questions and if they are giving useful answers. If you change any questions after a pre-test, you should not combine the results from the pre-test with the results of post-test interviews. Choosing sensible questions and ad ministering surveys with sensitivity and common sense will improve the quality of your results dramatically.

Data Collection

As a result of the pretest there will be observed some possible polarizations of the answers, which can be easily corrected. The pilot poll has got also the purpose to identify the structuring criteria, if these can not be identified. Step 10: The drawing up of the guide and the instructions to use the questionnaire. When a questionnaire is finalized, it has to be multiplied and distributed to the interview operators together with the Guide of usage. The guide solves the possible problems, which may appear during the interviews, and offers possible solutions for solving them, as for example: 1. Obtaining the identification information of the firm. The managers can be reserved to answering to questions about turnover, expenses, profit and if are not sure of the confidentiality of their data and the utility of the poll for the company. 2. Avoidance of powerful words. Consider the following example: at the question should all employees be forced by the trade-union to participate at improving courses so that increases the labor productivity with options of answering Yes or No. the answers will be a majority of No, even those answered by trade-union members. Words like promotion, dealer, base soft, application soft, and the computers send powerful signals to certain persons and can affect dramatically their answers. That is why it is important to keep in mind the influence of these terms on the results of the interview. 2.6.3 General Data Collection Tips Keep the questionnaire as short as possible. More people will complete a shorter questionnaire, regardless of the interviewing method. If a question is not necessary, do not include it. Put a title to your questionnaire (e.g., Leisure Activities Survey). Always include a short introduction - who you are and why you are doing the survey. This can be at the beginning of the questionnaire of could be as an introductory letter to the questionnaire. It is often a good idea to give the name of the research company rather than the client (e.g., XYZ Research Company rather than the producer of the product/service being surveyed).

Statistics for Business Administration

Include a cover letter with all mail surveys. A good cover letter will increase the response rate. A bad one or none at all, will reduce the response rate. Include the information in the preceding two paragraphs and mention the incentive (if any). Describe how to return the questionnaire. Include the name and telephone number of someone the respondent can call if they have any questions. Include instructions on how to complete the survey itself. Reassure your respondent of the total confidentiality and anonymity of the answers Mail questionnaires should be numbered on each page and include the return address on the questionnaire itself, because pages and envelopes can be separated from each other. Envelopes should have return postage prepaid. Using a postage stamp often increases response rates, but increases costs Optional you should give the possibility for the respondent to put the name, title, or phone number, if wish for. Some people will put in their names, making it possible for you to contact them again for clarification or follow-up questions. Do not ask for names, or phone number if the questions are sensitive in nature. Some people would become suspicious and not complete the survey. If the survey contains commercially sensitive material, ask a "security" question up front to find whether the respondent or any member of his family, household or any close friend works in the industry being surveyed. If so, terminate the interview immediately. The best way to ask security questions is in reverse (i.e., if you are surveying for a pharmaceutical product, phrase the question as "We want to interview people in security system industry - do you or any member of your household work in this area?). If the answer is "Yes" thanks the respondent and ends the interview. After the security question, start with general questions. If you want to limit the survey to users of a particular product, you may want to disguise the qualifying product. As a rule, start from general attitudes to the class of products, through brand awareness, purchase patterns, specific product usage to questions on specific problems (i.e., work from "What types of yoghourt have you bought in the last three months" to "Do you recall seeing a special offer on your last purchase of Brand X of yoghourt?"). If possible put the most important questions into the first half of the survey.

Data Collection

If a person gives up half way through, at least you have the most important information. Comprise all the relevant alternatives as answer choices. Leaving out a choice can give misleading results. For example, a number of recent polls that ask Americans if they support the death penalty yes or no have found 70-75% of the respondents choosing Yes. But polls that offer the choice between the death penalty and life in prison without the possibility of parole show support for the death penalty at about 50-60%. While polls that offer the alternatives of the death penalty or life in prison without the possibility of parole, with the inmates working in prison to pay restitution to their victims families have found support of the death penalty closer to 30%. Do not put two questions into one. Avoid questions such as "Do you buy frozen meat and frozen fish?" A "Yes" answer can mean the respondent buys meat or fish or both. Similarly with a question such as "Have you ever bought Product X and, if so, did you like it?" A "No" answer can mean "never bought" or "bought and disliked." Be as specific as possible. "Do you ever had sodas?" can include someone who once bought some in 1990. It does not tell you whether the sodas was with or without bubbles and may include someone who had soda in a restaurant. It is better to ask: "Have you bought sodas (other than in a restaurant) in the last three months?". "If yes, was it with or without bubbles?" Few people can remember what they bought more than three months ago unless it was a major purchase such as an automobile or a house equipment. Avoid emotionally charged words or leading questions that point towards a certain answer. You will get different answers from asking "What do you think of the XYZ proposal?" than from "What do you think of the Communist XYZ proposal?" The word "Communist" in the second question would cause some people to favor or oppose the proposal based on their feelings about Communism, rather than about the proposal itself. It is very easy to create bias in a questionnaire. This is another good reason to test it before going ahead. If you are comparing different products to find preferences, give each one a neutral name or reference. Do not call one "A" and the second one "B." This immediately brings images of A grades and B grades to mind, with the former being seen as superior to the latter. It is better to give each a

Statistics for Business Administration

"neutral" reference such "M" or "N" that do not have as strong a quality difference image. Avoid technical terms and acronyms, unless you are absolutely sure that respondents know they mean. LAUTRO, AGI, GPA, EIEIO (Life Assurance and Unit Trust Regulatory Organization, Adjusted Gross Income, Grade Point Average and Engineering Information External Inquiries Officer) are all well-known acronyms to people in those particular fields, but very few people would understand all of them. If you must use an acronym, spell it out the first time it is used. Make sure your questions accept all the possible answers. A question like "Do you use regular or premium gas in your car?" does not cover all possible answers. The owner may alternate between both types. The question also ignores the possibility of diesel or electric-powered cars. A better way of asking this question would be "Which type(s) of fuel do you use in your cars?" The responses allowed might be: a. Regular gasoline b. Premium gasoline c. Diesel d. Other e. Do not have a car If you want only one answer from each person, ensure that the options are mutually exclusive. For example: In which of the following do you live? a. A house b. An apartment c. The suburbs This question ignores the possibility of someone living in a house or an apartment in the suburbs.

Data Collection

Score or rating scale questions (e.g., "If '5' means very good and '1' means very poor how would rate this product?") are a particular problem. Researchers are extremely divided on this issue. Many surveys use a tenpoint scale, but there is considerable evidence to suggest that anything over a five-point scale is irrelevant. This depends partially on education. Among university graduates a ten-point scale will work well. Among people with less than a high school education five points is sufficient. In third world countries, a three-point scale (good/acceptable/bad) may be all some respondents can understand. Another problem is that you are assuming that the difference in the factors is within the scale limits - you may have a five-point scale but in a respondent's mind one factor may rate 10 points in comparison to the others. Another issue on which researchers differ is whether to use a scale with an odd or even number of points. Some like to force people to give an answer that is clearly positive or negative. This can make the analysis easier. Others feel it is important to offer a neutral, middle option. Your interviewing mode can make a difference here. A good interviewer can often get a answer, but in a self-administered interview, such as a Web page survey, a person who is frustrated by being unable to give a middle answer may leave a question blank, quit the survey, or both. Be sure any rating scale labels are meaningful. For example: What do you think about product X? a. It's the best on the market b. It's about average c. It's the worst on the market A question phrased like the one above will force most answers into the middle category, resulting in very little usable information. If you have used a particular scale before and need to compare results, use the same scale. Four on a five-point scale is not equivalent to eight on a ten-point scale. Someone who rates an item "4" on a five-point scale might rate that item anywhere between "6" and "9" on a ten-point scale. Be aware of cultural factors. In the third world, respondents have a strong tendency to exaggerate answers. Researchers may be perceived as being government agents, with the power to punish or reward according to the

Statistics for Business Administration

answer given. Accordingly they often give "correct" answers rather than what they really believe. Even when the questions are not overtly political and deal purely with commercial products or services, the desire not to disappoint important visitors with answers that may be considered negative may lead to exaggerated scores. People sometimes give answers they feel will reflect well on them. This is a constant problem for pre-election polls. More people say they will vote than actually will vote. More people say they go to museums or libraries than actually do. This problem is most significant when your respondents are talking directly to a person. People give more honest answers when answering questions on a computer. Mail surveys are in-between. In personal interviews it is vital for the operator to have empathy with the respondent. In general, operators should try to "mix" with respondents in terms of race, language, sex, age, etc. Choose your Interviewers according to the likely respondents. Leave your demographic questions (age, gender, income, education, etc.) until the end of the questionnaire. By then the operator should have built a rapport with the respondent that will allow honest responses to such personal questions. Mail questionnaires should do the same, although the rapport must be built by good question design, rather than personality. Exceptions are any demographic questions that qualify someone to be included in the survey. Do not have an interviewer ask a respondent's gender, unless they really cannot figure. Have the interviewer fill in the answer themselves. Paper questionnaires requiring text answers, should always leave sufficient space for handwritten answers. Provide a pen or a pencil Leave a space at the end of a questionnaire entitled "Other Comments." Always consider increase the attractiveness of the layout of your questionnaire. Make it attractive, easy to understand and easy to complete.

Surveys are a mixture of science and art, and a good researcher will save their cost many times over by knowing how to ask the correct questions.

Data Collection

2.7 Exercises Multiple choices exercices with answers 1. When data are collected in a statistical study for only a portion or subset of all elements of interest we are using: a. a sample b. a parameter c. a population d. a both b and c ANSWER: a 2. Which of the following must be avoided in designing a questionnaire? a. Dichotomous questions. b. Leading questions. c. Open-ended questions. d. Demographic questions. ANSWER: b 3. A questionnaire comprising particularly open questions is used mainly as a data collection support for: a. a formal undisguised study b. an informal, undisguised study c. a formal, disguised study d. an informal, disguised study e. all the above mentioned choices ANSWER: b 4. Mail data collection has as a main disadvantage is: a. a high rate of non-response b. a high cost c. a large number of interview operators d. low accuracy e. has no disadvantage ANSWER: a

Statistics for Business Administration

Multiple choice exercises without answers 5. a. b. c. d. e. 6. a. b. c. d. e. An overall research does not offer information concerning : the current status of the population all the population elements any sample the population the future status of the population Any statistical research will necessarily start with : data collection defining the population to be studied establishing the research plan establishing the research purpose developing the research plan

7. Into a formal interview, clear, undisguised interview: a. the operator should know the questions by hearth and will lead the discussion according the interviewed persons answers b. the operator marks the answers chosen by the interviewed person c. the interviewed person may chose even other answers as the ones comprised in the questionnaire d. the interviewed person can give free answers to the questions e. all the above variants are possible to occur 8. Data control and evaluation represent the activity to be conducted: a. previous to data summarizing b. before data collection c. before and after the data collection d. after the research report e. never

You might also like