Abstract

The results of a national fear of crime survey are compared with results following the use of different nonresponse correction procedures. We compared naive estimates, weighted estimates, estimates after thorough nonresponse follow-up and estimates after multiple imputation (MI). A strong similarity between the MI- estimates and the follow-up estimates was found. This suggests, that if the assumption of the data missing at random (MAR) holds, carefully selected and collected additional data used in MI procedures could yield similar estimates to a nonresponse follow-up at a much lower price and respondent burden.