Two Attempts to Increase the Response to a Mail Survey

Two measures intended to increase response were tested in a mail survey of members of the New Zealand general public. The first involved an envelope message designed to establish the non-commercial status of the survey and emphasise its importance. The treatment had the opposite effect to that expected. The second measure involved the use of a graphic design to create a more likeable questionnaire cover. The treatment had no effect on response. Both measures tested were identified by previous mail survey non-respondents as influencers of survey response, but these results emphasise the fallibility of survey respondents as predictors of their own or other respondents' behaviour. [PUBLICATION ABSTRACT]