This project aims to produce several contributions of considerable policy and practical relevance, as well as of methodological interest, for the study of the returns to educational qualifications using data from the National Child Development Survey. More precisely, the aim of our research is threefold.

First, we will derive estimates of the returns for the UK that allow for the possibility of misreported attainment. To date, any major empirical evidence on the importance of this issue is restricted to the United States, where it was shown that errors might indeed play a non-negligible role. We aim to provide the academic and policy community with estimates of the accuracy of commonly used types of data on educational qualifications for the UK: administrative files, self-reported information close to the date of completion of the qualification and recall information ten years after completion.

Second, by using the unique nature of our data we will assess how the biases from measurement error and from omitted ability and family background variables interact in the estimation of returns. We aim to produce simple calibration rules to allow policymakers to use nationally representative data, such as the Labour Force Survey, that totally rely on self-reported information on the qualifications and often contain little or no information on individual ability and family background.

Third, we will develop a unified framework of general applicability to study the impact of misreported treatment status on the estimation of the causal effects of a programme. Our discussion will thus be relevant for various areas of micro-econometric research.