Interviewer and Respondent Sociodemographic Characteristics, Rapport, and Their Joint Impact on Data Quality in the NESDA Study
Melany Horsfall, Adriaan W. Hoogendoorn, Stasja Draisma, Merijn Eikelenboom, Brenda W. J. H. PenninxABSTRACT
Objectives
Data quality in face‐to‐face interviews may be affected by sociodemographic characteristics of interviewers and respondents (age, education, and gender) as well as their (dis)similarity. Similarity in these characteristics may foster rapport, potentially enhancing cooperation and improving data quality. This study examined the associations between interviewer and respondent sociodemographics, their similarity, and data quality, and whether rapport mediates these relationships in a sample of respondents with and without a depressive or anxiety disorder.
Methods
Baseline data from the Netherlands Study of Depression and Anxiety were analyzed. Data quality was measured using item nonresponse and socially desirable responses to sensitive questions; rapport was assessed via post‐interview evaluations from both parties. Multilevel regression and mediation analyses were conducted.
Results
Rapport was lower among older respondents but higher with older interviewers. More dissimilarity in age was negatively associated with rapport sum score and dissimilarity in education was negatively associated with rapport scores from interviewers. Furthermore, older respondents had higher item nonresponse and, along with male respondents and those with higher education, gave more socially desirable responses. Rapport partially mediated the effect of respondent age on data quality. Additionally, having a depressive or anxiety disorder was negatively associated with rapport. Experiencing a depression was also linked to more item nonresponse, in contrast, having a depressive disorder or anxiety leads to less socially desirable responses.
Conclusion
These findings highlight the importance of interviewer training in rapport‐building, particularly with older respondents and respondents who suffer from a psychiatric disorder, to enhance data quality.