摘要：The objective of this study was to provide an estimate of the number of wrongly addressed envelopes that will be returned in a New Zealand mail survey. A total of 200 households were selected from throughout the country from regional telephone directories. Each envelope contained a covering letter, a questionnaire on home heating, and a reply paid envelope. The name, Miss J P Nettlefold, was used as the fictitious addressee. To the researcher's knowledge, none of the sample households contained a person with this name or anything resembling it. Of the 200 questionnaires posted, 137, or nearly 70%, were returned. A return rate of 65% was achieved for the 180 questionnaires posted to non-rural delivery addresses. The results of this study suggest that the number of envelopes returned marked as "Gone - no address" or "Return to sender" should be multiplied by about 1.5 to provide an estimate of the total number of envelopes incorrectly addressed. The results also suggest that approximately five per cent of the total wrongly addressed questionnaires will be returned completed to the researcher, at least for surveys of this type
关键词：mail survey; response rates; non response bias
