The impact of incentives on the response rates for FAMEX 1990 : an evaluation / S. Kumar and A. Durning. : CS11-613/92-1E-PDF

"It is well-known that decreases in response rates generally have an adverse effect on the quality of the resulting estimates. In an attempt to improve response rates for the 1990 Family Expenditure Survey, incentives were given to part of the sample. An experiment, evaluating the impact of the incentives, was undertaken. Under the design, some of the sampled households were given one of two incentives. This paper describes the analysis undertaken to evaluate the impact of incentives on response rates. For each city in the sample, the effects of the incentives are investigated using Pearson's chi-square test. Additional analysis at the national level is undertaken, and it is based on the sign (not the magnitude) of differences in the response rates between the incentive and non-incentive (control) groups. Further analysis, using Wald tests for logit models, examines not only the effects of the incentives on the response rates, but the effects of interviewer experience and regional differences as well. Pairwise comparisons are also used to investigate effects that are found to be significant in earlier analysis"--Abstract.

Lien permanent pour cette publication :
publications.gc.ca/pub?id=9.837231&sl=1

• Format MARC XML
• Format MARC HTML

Demander des formats alternatifs