Many people are unwilling or unable to provide investigations with insight into their conscious or unconscious attitudes. Disguised methods usually referred to as projective techniques have been developed to overcome this problem. Projective techniques are vague or incomplete stimuli to which the respondent is asked to respond. In doing so, it is believed that respondents reveal elements of their attitudes that they would not reveal in responses to direct questions. Unconscious attitudes that a respondent cannot verbalize or conscious attitudes that a respondent will not report because they are socially unacceptable may be explored by the projective method.
Projective techniques suffer from the same kinds of weaknesses as do depth interviews. Some psychologists question their validity but studies have shown that such techniques can provide more valid data than direct questions.
There are a variety of projective techniques, including word association, sentence completion story completion and pictorial tests. Each of these will be discussed briefly.
Word Association: Word association is one of the oldest and simplest projective techniques. Respondents are presented with a number of different words, one at a time. After each word they are asked to give the first word that comes to mind. If the list of words presented is related to the subject of interest, respondents may indicate some of their attitudes toward the subject with their responses. The underlying assumption is that by free associating with certain words, respondents will reveal their inner feelings about the subject being studied. Responses are timed so that those answers that respondents ‘reason out’ are identified and taken into account in the analysis.
Word association tests are not difficult to administer, for to most respondents taking the test is like laying a game. Nor is it difficult to construct a word list. It does require however some skill and experience to interpret the results. The usual way of constructing such a test is for researchers to prepare a list composed of a combination of stimulating and neutral words. Which stimulating words to include depends on the study purpose. For example, the purpose is to find responses to alternative advertising appeals, then the key words in these appeals will certainly be included in the list. On the other hand if one is undertaking an exploratory study on say consumers’ feelings about a particular food, then a great variety of words that might relate to food in general as well as that specific type of food would be used.
The words are usually read to the respondent one at a time, and the interviewer records the first word associated. Respondents should not be asked to write their answers because an additional variable is thereby added namely the delay required for the respondent to put the association into written form.
There are several variations to the simple test situation described above. For example, respondents may be asked to give not only the first word that comes to mind, but the first three or four. Variations of this technique may be run as controlled tests, as contrasted to free association. For example respondents may be asked what brand of cake mix comes to your mind first when I mention a moist cake? or what brand of detergent comes to your mind first when I mention soft and fluffy clothes? Another variation would be to ask. Who would eat a lot of oat-meal? Who a little?
A movie star
In analyzing the results of word association tests, the usual practice is to arrange responses along such liens as favorable non-favorable and pleasant unpleasant Individual questionnaires should be checked for consistency and for evidence of blocks. The latter are usually indicated by a person’s inability to associate within a time limit such as five seconds. This is why interviewers time respondents and proceed to the next word if no answer is forthcoming.