Affiliations: [a] Data Research, Access, and Governance Network, University of the West of England, Bristol, UK | [b] Social Research Institute, University College London, London, UK | [c] Bayes Business School, City, University of London, London, UK | [d] National Institute of Economic and Social Research, London, UK | [e] University of Reading, Reading, UK
Abstract: The role of the National Statistical Institution (NSI) is changing, with many now making microdata available to researchers through secure research environments This provides NSIs with an opportunity to benefit from the methodological input from researchers who challenge the data in new ways This article uses the United Kingdom’s Annual Survey of Hours and Earnings (ASHE) to illustrate the point We study whether the use of prefilled forms in ASHE may create inaccurate values in one of the key fields, workplace location, despite there being no direct evidence of it in the data supplied to researchers. We link surveys to examine the hypothesis that employees working for multi-site employers making an ASHE survey submission are more likely to have their work location incorrectly recorded as the respondent fails to correct the work location variable that has been pre-filled. In the short-term, suggestions are made to improve the quality of ASHE microdata, while longer-term we suggest that the burden of collecting additional data could be offset through greater use of electronic data capture. More generally, in a time when statistical budgets are under pressure, this study encourages NSIs to make greater use of the microdata research community to help inform statistical developments.
