In a survey for a trucking firm, traffic managers of manufacturing plants and department stores were shown a map of the United States with a section equivalent to about eight states outlined in red. The traffic managers were then asked: What annual volume in tons do you ship to or from the area outlined in red on the map? It is likely that most of the traffic managers had this information, but it is also likely that most of them would have had to compile the data from written records to give a very accurate report. Few respondents will take a lot of time to collect the information needed to answers a questionnaire unless they have a particular interest in the subject or believe they will benefit from the research. Otherwise they may make the best guess they can. If it is a mail questionnaire, they are apt to drop the questionnaire in the wastebasket. Such questions even if answered tend to irritate respondents and hence, damage cooperation with to rest of the survey. In most instances such questions should be avoided.
Will Respondents Give the information? Even though they know the answer, respondents will sometimes not answer question. There are two reasons for this: (1) they are unable to phrase their answers, or (2) they do not want to answer. It goes without saying that such questions hurt cooperation for the rest of an interview and, therefore should be eliminated. It is often possible, to change such questions to secure the desired information.
General Motors developed a classic questionnaire to get consumer opinions on car styling. Few consumers could describe the type of car grillwork that they liked best, but if they were shown pictures of 10 cars they could point out the grill they liked best. The pictures made it possible for respondents to report their answers.
Even though they could answer accurately, respondents may hesitate to do so if the question is seen as embarrassing or as relating to their social status. A questionnaire on magazine readership, for example showed relatively little readership of pulp magazines in the upper income groups. On the theory that these people mighty feel it was degrading to admit reading such ‘trash’, the study was revised. People were asked to contribute their old magazines to charity. A count of these magazines showed that the number of people contributing certain magazines exceeded the number who reported reading them by 50 percent.
When asked on what matters they were least willing to be interviewed, one group of respondents reported money matters, family life, political beliefs and religious beliefs in that order. Some income classification however, is desirable on many research projects. Other embarrassing or status damaging questions may be necessary for particular surveys. Numerous techniques, of which the following are examples, have been developed to encourage respondents to answer questions they may otherwise hesitate to answer.
1. The question may be included in a group of others that are more innocuous and the whole list asked quickly.
2. The interviewer may make a statement indicating that the behavior in question is not unusual and then specific question of the respondents. The statement tends to make respondents feel that their own behavior is not out of place.
3. The question may be phrased to refer to ‘other people’. For example, the respondent may be asked of most people report errors on their bills that are favorable to themselves. They will presumably answer in terms of their own practice.
4. A special ballot may be provided that respondents can complete personally and drop in a sealed box. This can be used only in personal interviews.
5. Another approach that can be used in personal interviews is that of handling respondent cards with alternatives listed and identified by letters or numbers. They can then respond in terms of the letter or umber. For example, the interviewer may hand out cards on which the following information is entered and ask respondents in which class their annual income falls:
Class L less than $20,000
Class M $20,000 — $40,000
Class N more than $40,000
It may be desirable to have enough classes so that both extremes are so extreme that few respondents will be in those classes.