Abstract: | Using data from a 1987 telephone survey of over 1,000 Illinoisadult residents, this research looked into factors causing maleunder-representation in surveys using the last-birthday methodof respondent selection. Differences by sex in rates of refusaland non-contact (after respondent selection) were evaluatedfor relative contribution to gender bias. Accuracy of informant'sresponse to the birthday question was gauged by checking againstbirth dates of adults rostered during the interview. Other factors,such as the proportion of informants who do not know the lastbirthday, or who misreport from a desire to be the one interviewed(or not interviewed), were evaluated for relative effect onsample representation of males and females. |