Reporting and measuring English school qualifications: a case study of General Certificate of Secondary Education results in survey and linked administrative data in the UK Millennium Cohort Study

Data on educational qualifications is essential in many research domains. The UK Millennium Cohort Study collected self-reported General Certificate of Secondary Education (GCSE) data in sweep 7 (cohort members aged 17). GCSE data from the National Pupil Database (NPD) has been linked to the MCS. This study investigates the consistency of these two sources of GCSE data using detailed statistical analyses and provides a concise sensitivity analysis to compare results obtained using the alternative data sources and measures. The authors observe marked inconsistencies between MCS self-reported GCSE results and the GCSE results in the NPD. Their central recommendation is that researchers should use the administrative GCSE records from the NPD whenever it is practicable. An open and transparent resource containing research code to process GCSE data from the MCS and from the NPD has been made available alongside this article.