Completion and Response Rates for Different Forms of Income Questions in a Mail Survey

In a mail survey, this experiment tested the hypotheses that return and completion rates to income questions could be increased by: (1) using wide instead of narrow categories and (2) reducing reactance by offering the respondent freedom of choice in reporting income. Neither hypothesis was supported. The results suggest that narrow categories be used if more detail on income would better fit the research objectives. In contrast to earlier research, this study questions the applicability of reactance theory to explaining responses to sensitive questions in a mail survey.