Use of a coded postcard to maintain anonymity in a highly sensitive mail survey: cost, response rates, and bias

Abstract

In a survey about euthanasia, 1,600 critical care nurses were randomly assigned to receive either three complete, anonymous mailings of the questionnaire or, with each mailing, a coded postcard to be returned separately from the questionnaire to reduce subsequent mailings to previous responders. The response rate in these two groups was 76.5% [95% confidence interval (CI) = 73.6-79.4%] and 69% (95% CI = 65.7-72.4%), respectively. The two strategies yielded similar responses, and costs were much lower for the postcard group. Using coded postcards to be returned separately from completed instruments appears to lower the response rate to anonymous mail surveys, but it also lowers cost and may not introduce additional bias.