Evaluating the Utility of Indirectly Linked Federal Administrative Records for Nonresponse Bias Adjustment

Sakshaug, Joseph W.; Antoni, Manfred

doi:10.1093/jssam/smy009

Cited by 9 publications

(13 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Exploration of this topic, however, is mainly limited to cross-sectional studies. Sakshaug and Antoni (2019) found promising correlation patterns between linked administrative variables and the response and survey variables, but adding these auxiliary variables into the nonresponse weighting scheme had only minor impact on survey estimates. Bee et al (2015) report mixed evidence for nonresponse bias adjustment depending on the variable of interest using CPS data enriched by tax records.…”

Section: Linked Administrative Data For Attrition Adjustment In a Lon...mentioning

confidence: 99%

“…One oftenused auxiliary data source is paradata -data about the survey process (Kreuter 2013), such as call record data or interviewer observations of the household/neighborhood. While these data are moderately associated with response propensities, their associations with survey variables are rather weak (Lin and Schaeffer 1995;Kreuter et al 2010b;Kreuter and Kohler 2009;Sakshaug and Antoni 2019;West et al 2014). Another possibility is to link commercial data to the sampling frame.…”

Section: Auxiliary Data Sources For Nonresponse and Attrition Adjustmentmentioning

confidence: 99%

See 1 more Smart Citation

Evaluating the Utility of Linked Administrative Data for Nonresponse Bias Adjustment in a Piggyback Longitudinal Survey

Büttner

Sakshaug

Vicari

2021

Journal of Official Statistics

View full text Add to dashboard Cite

Nearly all panel surveys suffer from unit nonresponse and the risk of nonresponse bias. Just as the analytic value of panel surveys increase with their length, so does cumulative attrition, which can adversely affect the representativeness of the resulting survey estimates. Auxiliary data can be useful for monitoring and adjusting for attrition bias, but traditional auxiliary sources have known limitations. We investigate the utility of linked-administrative data to adjust for attrition bias in a standard piggyback longitudinal design, where respondents from a preceding general population cross-sectional survey, which included a data linkage request, were recruited for a subsequent longitudinal survey. Using the linked-administrative data from the preceding survey, we estimate attrition biases for the first eight study waves of the longitudinal survey and investigate whether an augmented weighting scheme that incorporates the linked-administrative data reduces attrition biases. We find that adding the administrative information to the weighting scheme generally leads to a modest reduction in attrition bias compared to a standard weighting procedure and, in some cases, reduces variation in the point estimates. We conclude with a discussion of these results and remark on the practical implications of incorporating linked-administrative data in piggyback longitudinal designs.

show abstract

Section: Linked Administrative Data For Attrition Adjustment In a Lon...mentioning

confidence: 99%

Section: Auxiliary Data Sources For Nonresponse and Attrition Adjustmentmentioning

confidence: 99%

Evaluating the Utility of Linked Administrative Data for Nonresponse Bias Adjustment in a Piggyback Longitudinal Survey

Büttner

Sakshaug

Vicari

2021

Journal of Official Statistics

View full text Add to dashboard Cite

show abstract

“…In practice, the availability of auxiliary data for both respondents and nonrespondents is typically limited, leaving researchers little choice over which variables to use to build indicators of nonresponse bias (or nonresponse adjustment weights) -especially in the context of cross-sectional surveys (e.g. Sakshaug and Antoni 2018). If auxiliary data do exist, they typically consist of socio-demographic variables (e.g.…”

Section: Using R-indicators To Detect Nonresponse Bias -The Role Of Amentioning

confidence: 99%

“…variables that correlate both with the probability of responding and key survey variables), which (in cross-sectional surveys at least) is often limited (e.g. Sakshaug and Antoni 2018). Given that R-indicators essentially summarise nonresponse bias in the auxiliary variables, the question is raised as to how effective they (and the auxiliary variables) are at identifying the risk of nonresponse bias on other survey variables -especially in the context of large-scale surveys covering a wide variety of topics.…”

Section: Introductionmentioning

confidence: 99%

A Validation of R-Indicators as a Measure of the Risk of Bias using Data from a Nonresponse Follow-Up Survey

Roberts

Vandenplas²,

Herzing

2020

Journal of Official Statistics

View full text Add to dashboard Cite

R-indicators are increasingly used as nonresponse bias indicators. However, their effectiveness depends on the auxiliary data used in their estimation. Because of this, it is not always clear for practitioners what the magnitude of the R-indicator implies for bias in other survey variables, or how adjustment on auxiliary variables will affect it. In this article, we investigate these potential limitations of R-indicators in a case study using data from the Swiss European Social Survey (ESS5), which included a nonresponse follow-up (NRFU) survey. First, we analyse correlations between estimated response propensities based on auxiliary data from the register-based sampling frame, and responses to survey questions also included in the NRFU. We then examine how these relate to bias detected by the NRFU, before and after adjustment, and to predictions of the risk of bias provided by the R-indicator. While the results lend support for the utility of R-indicators as summary statistics of bias risk, they suggest a need for caution in their interpretation. Even where auxiliary variables are correlated with target variables, more bias in the former (resulting in a larger R-indicator) does not automatically imply more bias in the latter, nor does adjustment on the former necessarily reduce bias in the latter.

show abstract

“…Data quality has always been of great importance to the RDC-IAB, which is why it conducts methodological research on data quality, e.g., by using linked data in validation studies [12,13] or in analyses of survey nonresponse [14]. Data users also contribute to improving data quality, often by publishing data-related results in the FDZ-Methodenreport series.…”

Section: Developmentmentioning

confidence: 99%

The Research Data Centre of the German Federal Employment Agency at the Institute for Employment Research (RDC-IAB)

Antoni

Schmucker

2019

IJPDS

Self Cite

View full text Add to dashboard Cite

Since 2004, the Research Data Centre of the German Federal Employment Agency at the Institute for Employment Research (RDC-IAB) has been offering comprehensive individual data on employees, unemployed persons, job seekers and participants in active labour market programmes for scientific labour market research. For this purpose, data from employer notifications and from different administrative processes in the labour market administration are linked. These administrative data are also combined with survey data. In addition, linked employer-employee data allow simultaneous analyses of the supply and demand sides of the labour market. The data can be linked using unique identifiers, such as social insurance number, client number of the local employment agency, or establishment number. Since the foundation of the German Record Linkage Center (GRLC) in 2011, the RDC-IAB also applies methods for linking with non-unique and error-prone linkage identifiers like names, addresses and birth dates. German data protection law classifies the data offered by the RDC-IAB as highly sensitive and strictly regulates their use by external researchers. The RDC-IAB has therefore established various data access modes. Although the data can be transferred directly to research institutions in anonymised form, this procedure is generally not effective for linked data, as the loss of information due to the necessary anonymisation would be too great. For this reason, the RDC-IAB focuses on the access modes on-site use and remote data execution. In cooperation with other data centres, RDC-IAB has therefore established on-site data access at currently 16 locations worldwide.

show abstract

Evaluating the Utility of Indirectly Linked Federal Administrative Records for Nonresponse Bias Adjustment

Cited by 9 publications

References 23 publications

Evaluating the Utility of Linked Administrative Data for Nonresponse Bias Adjustment in a Piggyback Longitudinal Survey

Evaluating the Utility of Linked Administrative Data for Nonresponse Bias Adjustment in a Piggyback Longitudinal Survey

A Validation of R-Indicators as a Measure of the Risk of Bias using Data from a Nonresponse Follow-Up Survey

The Research Data Centre of the German Federal Employment Agency at the Institute for Employment Research (RDC-IAB)

Contact Info

Product

Resources

About