Dataset Information

Interviewer effects in food acquisition surveys.

ABSTRACT:

Objective

To understand the effects of interviewers on the responses they collect for measures of food security, income and selected survey quality measures (i.e. discrepancy between reported Supplemental Nutrition Assistance Program (SNAP) status and administrative data, length of time between initial and final interview, and missing income data) in the US Department of Agriculture's National Household Food Acquisition and Purchase Survey (FoodAPS).

Design

Using data from FoodAPS, multilevel models with random interviewer effects were fitted to estimate the variance in each outcome measure arising from effects of the interviewers. Covariates describing each household's socio-economic status, demographics and experience in taking the survey, and interviewer-level experience were included as fixed effects. The variance components in the outcomes due to interviewers were estimated. Outlier interviewers were profiled.

Setting

Non-institutionalized households in the continental USA (April 2012-January 2013).

Subjects

Individuals (n 14 317) in 4826 households who responded to FoodAPS.

Results

There was a substantial amount of variability in the distributions of the outcomes examined (i.e. time between initial and final interview, reported values for food security, individual income, missing income) among the FoodAPS interviewers, even after accounting for the fixed effects of the household- and interviewer-level covariates and removing extreme outlier interviewers.

Conclusions

Interviewers may introduce error in food acquisition survey data when they are asked to interact with the respondents. Managers of future surveys with similarly complex data collection procedures could consider using multilevel models to adaptively identify and retrain interviewers who have extreme effects on data collection outcomes.

SUBMITTER: Ong AR

PROVIDER: S-EPMC10261034 | biostudies-literature | 2018 Jul

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Interviewer effects in food acquisition surveys.

Ong Ai Rene AR Hu Mengyao M West Brady T BT Kirlin John A JA

Public health nutrition 20180222 10

<h4>Objective</h4>To understand the effects of interviewers on the responses they collect for measures of food security, income and selected survey quality measures (i.e. discrepancy between reported Supplemental Nutrition Assistance Program (SNAP) status and administrative data, length of time between initial and final interview, and missing income data) in the US Department of Agriculture's National Household Food Acquisition and Purchase Survey (FoodAPS).<h4>Design</h4>Using data from FoodAPS ...[more]

PMID: 29467041

Similar Datasets

Project description:BackgroundA population-level survey (PLS) is an essential and standard method used in public health research that supports the quantification of sociodemographic events, public health policy development, and intervention designs. Data collection mechanisms in PLS seem to be a significant determinant in avoiding mistakes. Using electronic devices such as smartphones and tablet computers improves the quality and cost-effectiveness of public health surveys. However, there is a lack of systematic evidence to show the potential impact of electronic data collection tools on data quality and cost reduction in interviewer-administered surveys compared with the standard paper-based data collection system.ObjectiveThis systematic review aims to evaluate the impact of the interviewer-administered electronic data collection methods on data quality and cost reduction in PLS compared with traditional methods.MethodsWe conducted a systematic search of MEDLINE, CINAHL, PsycINFO, the Web of Science, EconLit, Cochrane CENTRAL, and CDSR to identify relevant studies from 2008 to 2018. We included randomized and nonrandomized studies that examined data quality and cost reduction outcomes, as well as usability, user experience, and usage parameters. In total, 2 independent authors screened the title and abstract, and extracted data from selected papers. A third author mediated any disagreements. The review authors used EndNote for deduplication and Rayyan for screening.ResultsOur search produced 3817 papers. After deduplication, we screened 2533 papers, and 14 fulfilled the inclusion criteria. None of the studies were randomized controlled trials; most had a quasi-experimental design, for example, comparative experimental evaluation studies nested on other ongoing cross-sectional surveys. A total of 4 comparative evaluations, 2 pre-post intervention comparative evaluations, 2 retrospective comparative evaluations, and 4 one-arm noncomparative studies were included. Meta-analysis was not possible because of the heterogeneity in study designs, types, study settings, and level of outcome measurements. Individual paper synthesis showed that electronic data collection systems provided good quality data and delivered faster compared with paper-based data collection systems. Only 2 studies linked cost and data quality outcomes to describe the cost-effectiveness of electronic data collection systems. Field data collectors reported that an electronic data collection system was a feasible, acceptable, and preferable tool for their work. Onsite data error prevention, fast data submission, and easy-to-handle devices were the comparative advantages offered by electronic data collection systems. Challenges during implementation included technical difficulties, accidental data loss, device theft, security concerns, power surges, and internet connection problems.ConclusionsAlthough evidence exists of the comparative advantages of electronic data collection compared with paper-based methods, the included studies were not methodologically rigorous enough to combine. More rigorous studies are needed to compare paper and electronic data collection systems in public health surveys considering data quality, work efficiency, and cost reduction.International registered report identifier (irrid)RR2-10.2196/10678.

Project description:BackgroundSelection bias in HIV prevalence estimates occurs if non-participation in testing is correlated with HIV status. Longitudinal data suggests that individuals who know or suspect they are HIV positive are less likely to participate in testing in HIV surveys, in which case methods to correct for missing data which are based on imputation and observed characteristics will produce biased results.MethodsThe identity of the HIV survey interviewer is typically associated with HIV testing participation, but is unlikely to be correlated with HIV status. Interviewer identity can thus be used as a selection variable allowing estimation of Heckman-type selection models. These models produce asymptotically unbiased HIV prevalence estimates, even when non-participation is correlated with unobserved characteristics, such as knowledge of HIV status. We introduce a new random effects method to these selection models which overcomes non-convergence caused by collinearity, small sample bias, and incorrect inference in existing approaches. Our method is easy to implement in standard statistical software, and allows the construction of bootstrapped standard errors which adjust for the fact that the relationship between testing and HIV status is uncertain and needs to be estimated.ResultsUsing nationally representative data from the Demographic and Health Surveys, we illustrate our approach with new point estimates and confidence intervals (CI) for HIV prevalence among men in Ghana (2003) and Zambia (2007). In Ghana, we find little evidence of selection bias as our selection model gives an HIV prevalence estimate of 1.4% (95% CI 1.2% - 1.6%), compared to 1.6% among those with a valid HIV test. In Zambia, our selection model gives an HIV prevalence estimate of 16.3% (95% CI 11.0% - 18.4%), compared to 12.1% among those with a valid HIV test. Therefore, those who decline to test in Zambia are found to be more likely to be HIV positive.ConclusionsOur approach corrects for selection bias in HIV prevalence estimates, is possible to implement even when HIV prevalence or non-participation is very high or very low, and provides a practical solution to account for both sampling and parameter uncertainty in the estimation of confidence intervals. The wide confidence intervals estimated in an example with high HIV prevalence indicate that it is difficult to correct statistically for the bias that may occur when a large proportion of people refuse to test.

Dataset Information

Interviewer effects in food acquisition surveys.

Objective

Design

Setting

Subjects

Results

Conclusions

Publications

Interviewer effects in food acquisition surveys.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets