A micro analysis of the macro differences in refusal risk among metropolitan areas. The case of the Italian multipurpose survey |
| |
Authors: | Romina Fraboni Alessandro Rosina Sante Orsini Barbara Baldazzi |
| |
Institution: | (1) Istituto Nazionale di Statistica (ISTAT), Via Adolfo Ravá, 150 Roma, Italy;(2) Universitá Cattolica del Sacro Cuore, Largo Gemelli 1, 20123 Milano, Italy |
| |
Abstract: | Although still modest, non response rates in multipurpose household surveys have recently increased, especially in some metropolitan
areas. Previous analyses have shown that refusal risk depends on the interviewers' characteristics. The aim of this paper
is to explain the difference in refusal risk among metropolitan areas by analysing the strategies adopted in the recruitment
of interviewers through a multilevel approach. The Annual Survey on Living conditions is a PAPI survey of the "Multipurpose"
integrated system of social surveys and it represents our data base. For non responding household, data on non response by
reason, municipality and characteristics of the interviewer are available. The results highlight that those cities recruiting
interviewers mainly among young students have a higher refusal risk. These results are particularly important as they indicate
that recruitment strategies may have a substantial impact on non sampling errors.
Acknowledgements An earlier version of this article was presented at the International Conference on Improving Survey, University of Copenhagen,
Denmark, August 25-28, 2002. We would like to thank the participants to the presentation for their useful comments and suggestions.
Opinions expressed are those of the authors and do not necessarily represent the official position of any of the institutions
they work for. |
| |
Keywords: | Non-response rate multilevel modelling interviewer's effect |
本文献已被 SpringerLink 等数据库收录! |
|