Part-Whole Question Order Effects: Views of Rurality

Abstract

A split-ballot design was used to assess part-whole question order effects in a mail survey. A general question asked for respondent's perceptions of overall life quality in rural areas relative to other areas; a series of 19 similar questions asked for relative quality of specific aspects of rural life. When general question was asked after specific items, subjects were more likely to answer general question, less likely to respond that rural areas were the same as other areas, and more posi- tive in their opinions about rural places. The addition of a pream- ble-instructing subjects to read both general and list of specific items before answering either-failed to alter these findings. The order effects on responses to general question did not appear to represent either a simple assimilation (carry- over) effect or a contrast (backfire) effect, both of which have been found by other researchers. Possible reasons for con- flicting findings are discussed. When related specific and general questions appear together, re- sponses to general question may differ, depending upon order in which questions are asked. In some cases, general question responses are contrasted with specific answers. Thus, Schuman and Presser (1981) found that respondents reported lower overall hap- piness when general item followed a specific question on marital happiness than when questions were reversed. Since most subjects reported high levels of marital happiness, they apparently responded to subsequent general question by subtracting their marital feelings as if general item had asked, Apart from your marriage, how happy are you with your life situation? Congruent with a contrast