Search results

A number of writers have attributed the ordinal position effect to the complexity or difficulty of the question and to the degree of crystallization of the attitude being investigated. This implies that a questionnaire on radio program preferences, presumably of little difficulty and involving relatively crystallized attitudes, should have no order effect. However, other factors such as nature of the R's or the use of check list questions alone rather than check lists buried among other types of questions also are hypothesized as bringing about an order effect. For example, an order effect which would not be found in student responses may appear in the responses of busy adults who feel imposed upon by the interviewer. A buried check list ? - 'Of the sixteen types of program material listed below, which FIVE do you like best?' - was used in 16,193 interviews of a stratified random sample of adults taken from Iowa & Kansas in 1952. 16 different forms were used, each program type appearing once on each form, once in each ordinal position, and preceding and following every other program type once. Analysis revealed signif (p.<.05) order effects for 8 of the 16 program types. Thus the later on a check list a program type appears, the less likely it will be chosen as one of the five favorites. Breakdown analysis for age, sex, and educ indicated no difference in the degree to which different R's were affected by order. K. Geiger.