DIUS commissioned this study to improve its understanding of the impact of lifelong learning on social mobility. The full programme of work considered four longitudinal data sets: the ONS Longitudinal Study (ONS-LS); the National Child Development Study (NCDS); the 1970 British Cohort Study (BCS70); and the British Household Panel Survey (BHPS). The research team led by Dr Peter Urwin at the University of Westminster, Dr Jo Blanden of the Centre for Economic Performance and Dr Patrick Sturgis of the University of Surrey.
As a modelling strategy we employed a finite distributed lag model with a fixed effects specification. A fixed effects specification enables us to isolate the role of education as distinct from other person-specific characteristics which might be correlated with wages and education level. If such factors are constant over time (for example parental occupation, or cognitive ability) then their influence is differenced out and will not affect the coefficient for changes in lifelong learning. This is particularly important if those who engage in lifelong learning possess unobserved characteristics which are themselves associated with higher earnings.