Interviewer gender and loneliness disclosure in survey research: evidence from the European Social Survey and the Survey of Health, Ageing and Retirement in Europe

Loneliness is an important public health concern, but interviewer effects in face-to-face surveys may affect population-based monitoring efforts. Prior work suggests that people may be more open about sensitive topics to female interviewers than to male interviewers. This study examines whether interviewer gender influences the reporting of loneliness. Data are from 276,073 face-to-face interviews from four rounds of the European Social Survey (ESS) and four waves of the Survey of Health, Ageing and Retirement in Europe (SHARE). Country-specific logistic regression models were estimated to assess the association between interviewer gender and reported loneliness, adjusting for respondent characteristics. Interaction effects with respondent gender were also examined. Country-specific average marginal effects were combined using random-effects meta-analyses. Sensitivity analyses used a continuous measure of loneliness (short UCLA scale). Across datasets, the effect of interviewer gender on reported loneliness was close to zero and not statistically significant (ESS: θ = −0.001, 95% CI: −0.006 to 0.005; SHARE: θ = −0.017, 95% CI: −0.034 to 0.001; Overall: θ = −0.004, 95% CI: −0.011 to 0.002). No evidence was found for stronger effects among male respondents (ESS: θ = 0.006; 95% CI: -0.006 to 0.019; SHARE: θ = -0.005; 95% CI: -0.027 to 0.018; Overall: θ = 0.002; 95% CI: -0.009 to 0.013). Although some between-country heterogeneity was observed, effect estimates remained centered around zero. Sensitivity analyses using the UCLA loneliness scale yielded comparable results. These precise estimates suggest that interviewer gender is unlikely to be a source of bias in loneliness monitoring.