Effects of Anonymity and Social Setting on Survey Responses

Anonymity was tested by placing an identifying number on halfof the questionnaires on a mail survey designed to elicit reactionsfrom teachers concerning teacher unions. Bias due to socialsetting was tested by mailing half the questionnaires to teachersat their home addresses, the other half to teachers at theirschool addresses. In a two-by-two factorial design, anonymityand social setting had no appreciable effect on response orspeed of return, although social setting did have an effecton return rate. Respondents receiving the questionnaire at theirschool address had a higher return rate than those receivingit at home