Whittard, D., Ritchie, F., Phan, V., Bryson, A., Forth, J., Stokes, L. and Singleton, C., 2023. The perils of pre-filling: lessons from the UK’s Annual Survey of Hours and Earning microdata. Statistical Journal of the IAOS, 39(3), pp.661-677.
This Research Article examines 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 it is suggested that the burden of collecting additional data could be offset through greater use of electronic data capture.
This Data Insight examines the incidence of low pay in Britain over the period 2004-2021 using two indicators that are the focus of government policy. The analyses use data developed as part of the Wage and Employment Dynamics project, funded by ADR England. We use the data to identify and adjust for biases in estimates of the incidence of low pay. We show that the incidence of minimum wage employment is under-estimated. We also show that the incidence of low pay has been falling faster than previously thought.
Forth, J; Phan, V; Stokes, L; Bryson, A; Ritchie, F; Whittard, D and Singleton, C (2022) Methodology Paper: Longitudinal Attrition in ASHE Interim Results: Attrition-paper-overview.pptx (live.com)
Abstract: The Annual Survey of Hours and Earnings (ASHE) provides many of the UK’s official earnings statistics. The survey operates on an annual 1% sample of employee jobs. However, the method of sampling – based on the final two digits of an employee’s National Insurance number – means that records are linkable longitudinally. Many government and academic studies have utilised the dataset in this way. However, the longitudinal integrity of the ASHE sample has been the subject of little prior investigation, with the panel sample generally assumed free of any attrition biases that might compromise longitudinal analysis. We explore the validity of this assumption by comparing rates of year-on-year sample retention in ASHE with rates of employment retention estimated from a reference dataset (the Longitudinal Annual Population Survey). Our analysis confirms the existence of systematic patterns of longitudinal attrition in ASHE, which have the potential to introduce bias into longitudinal analyses of these data. We go on to construct longitudinal weights that correct for estimated attrition biases over adjacent years in ASHE. In an illustrative analysis, the application of these weights brings about a small widening of the distribution of individual wage growth.
Whittard, D; Ritchie, F; Phan, V; Forth, J; Bryson, A; Stokes, L; Singleton, C; McKenzie, A (2022) Exploring the workplace location problem in the ASHE Interim Results: Workplace-location-overview.pptx (live.com)
Abstract: The Annual Survey of Hours and Earnings (ASHE) is an important source of longitudinal linked employer-employee payroll earnings data for Britain. It provides accurate information on employees’ hours and earnings and information on the location of employees and their place of work. This paper examines the accuracy of the workplace location in the data following concerns that the pre-filling of that data item by ONS in paper questionnaires results in measurement error. By linking the employee’s workplace in ASHE to the same organisations in the Business Structure Database (BSD), and by examining employee commuting distances, we confirm that there is systematic measurement error in the recording of the workplace location in the ASHE among organizations where the survey is administered via a paper questionnaire. The study suggests a number of alternative approaches to improve the quality of the data and reduce the potential for biased estimates. This is particularly important in the context of regional earnings differentials, with their attendant implications for the levelling-up policy agenda
Abstract: ASHE is a key dataset in the UK, the only one which allows long-term analysis of flows in labour market status and earnings, and hence vitally important in the understanding of low pay and wage progression. Separating out students from non-student workers therefore has considerable value. This study has tried to create a proxy for ‘student working’ using the ASHE dataset, and then triangulating with the Census 2011 data which has some of the same people but with an accurate marker for student status. Unfortunately, triangulating this with accurate student information on the Census suggested that our preferred method was not notably the ‘best’
Abstract: This feasibility study is part of the Wage and Employment Dynamics (WED) project, which is seeking to provide new insight into the dynamics of earnings and employment in the UK, by enhancing the Annual Survey of Hours and Earnings (ASHE) through linkage to other survey and administrative data. This feasibility study focused on the potential for linking education data. Specifically, it set out to: 1) Scope the possibilities for linking administrative education data to ASHE and, as far as possible, get a sense of any legal and practical barriers to pushing those linkages forward. 2) Summarise the types of research questions that could potentially be addressed by each linkage and try to provide some sense of which research questions are most pressing from a policy perspective. 3) Scope, as far as possible, any existing plans for other linkages which may deliver answers to some or all of these questions, and hence where any ASHE linkages could add most value. 4) Provide recommendations for which linkages should be taken forward and in what order
Abstract: The Annual Survey of Hours and Earnings (ASHE) is based on a 1% sample of employee jobs and provides many of the UK’s official earnings statistics. Weights are provided with the core dataset, which adjust the profile of the annual achieved sample such that it is representative of employees by gender, age, occupation and region. However, while the ASHE is based on a sample of employee jobs, the survey is completed by employers. Not all employers respond, and some do not respond quickly enough for their returns to be incorporated in the annual dataset. In this methodological note, we explore the characteristics of employers responding to ASHE. We show that certain types of organisations are over or under-represented in the achieved sample, even after applying the weights from the core dataset. We construct an adjustment to those existing weights that attempts to remove these biases. We discuss the implications for some headline estimates of earnings