The Payoff to Vocational Qualifications: Reconciling Estimates from Survey and Administrative Data – December 2017

Practice area: Economics of Education | Education and Labour Markets | Further education
Client: Centre for Vocational Education Research
Published: November, 2017
Keywords: quantitative analysis

Estimating the wage differentials associated with particular qualifications has increasingly become part of the general ‘returns to education’ literature. Two particular strands of the literature looking at such qualifications can be identified: the first uses data on individuals obtained through representative sample surveys, such as the Labour Force Survey (LFS) in the UK, while the second uses data from administrative sources, such as the Individualised Learner Record (ILR) in the UK, often matched to other datasets. Differing results have often been observed between them, particularly for low-level vocational qualifications, with the administrative data typically producing larger estimated differentials. In a joint report with the University of Sheffield we show that, when estimated on the same specification, survey and administrative data produce estimated earnings differentials that follow the same pattern in terms of relative size across qualifications, and that for most qualifications are similar in absolute size. There does not seem to be one particular specification difference that was driving the differing results in the literature, with some specification changes being more important for some qualifications, but less so for others. The full report is available from the Centre for Vocational Education Research website through the link above.