Finns’ self-selection into survey research as a function of researcher ethnicity: does signing the request letter with a foreign name (Russian or Arabic) attenuate response rates?

Le résumé fourni par la source

In this study, we investigated whether ethnic discrimination may influence participation in a typical traffic survey. An identical five-page survey was sent to a representative sample of Finnish adults (N = 2100); however, the ethnicity (Arab, Finnish, or Russian) of the sender was different for each of the three subsamples (N = 700). The response rate for a Finnish sender was 30.47%, while the Russian and Arab senders’ response rates were 26.68% and 25.75%, respectively (Wald χ2 = 4.369, d.f. = 2, p = 0.113). Collating the last two, we found that a non-Finnish name received less responses than a Finnish name (30.47% vs. 26.21%; Wald χ2 = 4.220, d.f. = 1, p = 0.040; OR (non-Finn vs. Finn) = 0.81, 95% CI: 0.66–0.99). Given that the effect size was small and was obtained in a post-hoc analysis, we conclude that we have not found strong evidence of ethnic discrimination in this low-stakes context.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.