On the Validity of Using Students as Subjects in Consumer Behavior Investigations

A problem of continuing concern to researchers in the area of consumer behavior is, When is it appropriate to use students as surrogates of the population that one is interested in examining? Because of time, distance, and money considerations, students are often used as quick and convenient sources of information. But are the results really generalizable to the population that the researcher is interested in investigating? Conflicting results have been noted so far. In a study on businessmen by Alpert, it was found that results from businessmen and those from students differed significantly.' The conclusion reached was that the best subjects for instruments designed to measure businessmen's behavior are businessmen themselves. Another study by Khera and Benson tends to support the position of the above findings-that the behavior of businessmen as subjects is likely to be a more accurate predictor of actual businessmen's behavior than is that of students.2 This study also suggests, however, that student subjects may be good substitutes for businessmen under certain conditions. In a dissonance-reduction investigation by Sheth, there was a remarkable degree of similarity between students' and housewives' responses.3 Another study on home economics majors indicated that students' answers to internally scaled questions generally had smaller variances than housewives' answers, and that students tended to give higher product ratings than the housewives.4 In a consumer behavior study being conducted with housewives by Enis, Cox, and Stafford, it was found in comparing student and housewife responses that significant differences at or at less than the .10 level were reported for five out of 11 product comparisons.' The results