Correlates of Record Linkage and Estimating Risks of Non-Linkage Biases in Business Data Sets

Abstract

Summary.Researchers often utilize data sets that link information from multiple sources, but non-linkage biases caused by linked and non-linked subject differences are little understood, especially in business data sets. We address these knowledge gaps by studying biases in linkable 2010 UK Small Business Survey data sets.We identify correlates of business linkage propensity, and also for the first time its components: consent to linkage and register identifier appendability. As well, we take a novel approach to evaluating non-linkage bias risks, by computing data set representativeness indicators (comparable, decomposable sample subset similarity measures). We find that the main impacts on linkage propensities and bias risks are due to consenternon-consenter differences explicable given business survey response processes, and differences between subjects with and without identifiers caused by register undercoverage of very small businesses. We then discuss consequences for the analysis of linked business data sets, and implications of the evaluation methods we introduce for linked data set producers and users.

Statistical Challenges of Administrative and Transaction Data

“…Stopping here would reduce the number of calls made by 7-15% and result in substantial financial savings. Moore et al (2018c) have studied biases in linkable 2010 UK Small Business Survey data sets. As well as assessing how representative a final linked data set would be, they have identified the correlates of consent to linkage, and whether a register identifier can be appended and hence a link made.…”

Improved Secondary Analysis of Linked Data: A Framework and an Illustration

“…In addition to the usual sources of survey bias (coverage bias, informative sampling bias and non-response bias), linked data are subject to a variety of potential additional biases, including consent bias (the subsample of survey respondents who consent to their data being linked can be non-representative), missed links bias (the subsample of consenters whose records can actually be linked can be non-representative) and incorrect links bias (caused by errors in the linkage process). See Harron (2016) and Moore et al (2018).…”

Using Linked Longitudinal Administrative Data to Identify Social Disadvantage

“…Consent bias may also be introduced when survey respondents are asked for permission to link administrative records to information collected in surveys. Individuals who provide consent to administrative data linkage may differ systematically from individuals who refuse consent and this can affect statistical inferences drawn from linked administrative data (Kho et al 2009;Moore et al 2018;Sakshaug et al 2012;Sala et al 2014).…”