A comparison of two population-based household surveys in Uganda for assessment of violence against youth

Violence is associated with health-risk behaviors, potentially contributing to gender-related HIV incidence disparities in sub-Saharan Africa. Previous research has demonstrated that violence, gender, and HIV are linked via complex mechanisms that may be direct, such as through forced sex, or indirect, such as an inability to negotiate safe sex. Accurately estimating violence prevalence and its association with HIV is critical in monitoring programmatic efforts to reduce both violence and HIV. We compared prevalence estimates of violence in youth aged 15–24 years from two Ugandan population-based cross-sectional household surveys (Uganda Violence Against Children Survey 2015 [VACS] and Uganda Population-based HIV Impact Assessment 2016–2017 [UPHIA]), stratified by gender. UPHIA violence estimates were consistently lower than VACS estimates, including lifetime physical violence, recent intimate partner physical violence, and lifetime sexual violence, likely reflecting underestimation of violence in UPHIA. Multiple factors likely contributed to these differences, including the survey objectives, interviewer training, and questionnaire structure. VACS may be better suited to estimate distal determinants of HIV acquisition for youth (including experience of violence) than UPHIA, which is crucial for monitoring progress toward HIV epidemic control.

Strategic value of interviewer training and local community-based organisations for WaSH and antibiotic resistance surveys

Water, sanitation, and hygiene (WaSH) research has long relied on household surveys to gather knowledge, attitudes, and practice (KAP) data with local enumerators or community correspondents (CCs). However, CCs must be trained to build capacity and ensure representative survey responses. Here, we use a case study in low-income, informal communities from New Delhi, India to assess the value of structured training for individuals who lead the gathering of KAP data (CC leaders, CCLs) on WaSH and antibiotic resistance. Feedback from CCLs showed that directed training increased their motivation, confidence, and technical competence and provided them skills that enhanced data collection. Training further strengthened relationships and empowered our local community-based organisation (CBO), expanding their role beyond being an implementing partner. Empowerment led to new insights, such as evident problems with communications between local doctors and community dwellers. Only three of 38 focus group attendees knew what an antibiotic was, apparently because they were never told. Overall, this work shows that interviewer training has many trickle-down benefits, improving the quality of data, building confidence in field teams, and empowering local CBOs, but most importantly, by increasing knowledge among community dwellers, such that they also might be empowered.

Do interviewer attitudes to data linkage influence respondents’ consent to linkage? Analysis of Understanding Society

Background Variable consent rates threaten the validity of linked datasets. One modifiable element is the interviewer–respondent relationship. We examine interviewer attitudes to consent to linkage and the effect on respondent consent. Methods Subjects were 27 380 respondents from the Wave 1 Understanding Society (US) survey in Great Britain and 449 interviewers who completed the US Interviewer Survey. Two types of consent were considered: (i) whether the interviewer would hypothetically agree to having their data linked if he/she was an US respondent and (ii) whether the respondent consented to have their data linked. Factors influencing the interviewer’s propensity to link data were examined using logistic regression. The association between interviewer consent and respondent consent to health record linkage was assessed using multi-level logistic regression models. Results The interviewer’s propensity to consent to data linkage was strongly positively associated with its perceived usefulness: those that found it somewhat useful were 57% less likely to consent [adjusted odds ratio (AOR) 0.43, 95% CI: 0.22–0.82] compared to those who thought it was very useful. Positive beliefs about data security and their ability to understand the data linkage information were also associated. Respondents were 17% less likely to consent when interviewed by an interviewer who would not consent to record linkage (AOR 0.83, 95% CI: 0.71–0.97). Conclusions The interviewer’s propensity to consent was influenced by their beliefs about data linkage, which in turn influenced respondent consent. We recommend using interviewer training to emphasize the usefulness of data linkage and the measures around data security.

How Well Do Interviewers Record Responses to Numeric, Interviewer Field-code, and Open-ended Narrative Questions in Telephone Surveys?

Telephone survey interviewers need to be able to accurately record answers to questions. While straightforward for closed questions, this task can be complicated for open questions. We examine interviewer recording accuracy rates from a national landline random digit dial telephone survey. We find that accuracy rates are over 90% for numeric response and interviewer-code, single-response items but are astonishingly low (49%) for a multiple-answer, nominal, interviewer-code item. Accuracy rates for narrative open questions were around 90% for themes but only about 70% for themes and elaborations. Interviewer behaviors (e.g., probing, feedback) are generally associated with lower accuracy rates. Implications for questionnaire design, interviewer training, and coding procedures are discussed.

What They Expect Is What You Get: The Role of Interviewer Expectations in Nonresponse to Income and Asset Questions

Personal income and assets are sensitive topics to discuss. This phenomenon is reflected in high rates of nonresponse to financial questions in surveys. In face-to-face surveys, item nonresponse is influenced by interviewers. Although interviewers are trained to conduct standardized interviews, some obtain a higher number of item nonresponses than others. This study examines interviewer effects on nonresponse to questions about household income, bank balances, and interest and dividend income in the Survey of Health, Ageing and Retirement in Europe (SHARE). It first investigates the extent to which interviewers affect nonresponse to income and asset questions and second whether interviewers’ prior expectations regarding respondents’ likelihood to provide information about their income predict actual nonresponse rates. Results of multilevel modeling show that interviewer influence on nonresponse to the income and asset questions was significant at the five percent level. In addition, interviewer expectations were significantly correlated with “don’t know” responses and “refusals.” These results indicate that interviewer expectations matter in the context of income and asset questions and that survey practitioners should take this into account when designing interviewer training.

The Effect of Question Characteristics on Question Reading Behaviors in Telephone Surveys

Asking questions fluently, exactly as worded, and at a reasonable pace is a fundamental part of a survey interviewer’s role. Doing so allows the question to be asked as intended by the researcher and may decrease the risk of measurement error and contribute to rapport. Despite the central importance placed on reading questions exactly as worded, interviewers commonly misread questions, and it is not always clear why. Thus, understanding the risk of measurement error requires understanding how different interviewers, respondents, and question features may trigger question reading problems. In this article, we evaluate the effects of question features on question asking behaviors, controlling for interviewer and respondent characteristics. We also examine how question asking behaviors are related to question-asking time. Using two nationally representative telephone surveys in the United States, we find that longer questions and questions with transition statements are less likely to be read exactly and fluently, that questions with higher reading levels and parentheticals are less likely to be read exactly across both surveys and that disfluent readings decrease as interviewers gain experience across the field period. Other question characteristics vary in their associations with the outcomes across the two surveys. We also find that inexact and disfluent question readings are longer, but read at a faster pace, than exact and fluent question reading. We conclude with implications for interviewer training and questionnaire design.