Sociologie Românească (Romanian Sociology), new series no. 3-4/2002, pp. 1-32.
O analiză a ratei de răspuns în anchetele de opinie naționale
An analysis of the response rate in national surveys
Abstract (Rezumat în limba engleză): The present analysis belongs to the wider context of sample representativeness and of the factors that modify actual sample reported to the project ones. In this article we explored the present state of the response rate in some opinion surveys lead in Romania and the factor that determine it. We identified four categories of factors, of the research and the context / environment. We used for analyses data gathered within several opinion survey conducted by Metro Media Transilvania during 2000-2002.
The main conclusion of the analyses were regarding the extent of the response rate, the factors influencing it, its evolution during the past 3 years, the motivation and the effect of replacements of the survey data. The response rate in the Romanian survey is rather low (60%), due mainly to the quality of the sampling frame (the election list) and not to the subject’ refusals (the refusal rate is rather low – 8%). Even if the selection of the interview operators is intentional, they still exert an influence on the response rate (higher response rates are obtain by the operators with an average experience, who do not have the residence in the locality were they apply the survey, who are interested in the research topic, having an under average scholar grades, and who wear few accesories). The response rate also depends on some characteristic of the research: it is higher in those surveys that are less imposing on the subject – shorter questionnaires, one-wave interviewing. It is also depends on some characteristics of the subjects, being higher in female subjects, from the rural area, more educated and more interested in the research topic.