Job Market Paper

The selection effect of early cognitive abilities on educational decisions

(with Andrew Jones & Nigel Rice)

Do educational decisions after compulsory education affect midlife cognitive ability? In this paper, we investigate the effect of three educational decisions on midlife cognitive ability (at age 46), after controlling for early cognitive abilities and socioeconomic factors, with data from the 1970 British Cohort Study. We use a structural equation modelling approach which includes a measurement model for latent abilities and a structural model to address the research question. We find that undergraduate and postgraduate education have a positive treatment effect on midlife cognitive ability. Since education levels are sequential, the total educational effect on midlife cognitive ability is cumulative, which indicates that higher the levels of education lead to higher midlife cognitive ability.

Work in Progress

The mediation effect of midlife cognitive ability on the midlife returns to educational decisions 

(with Andrew Jones & Nigel Rice)

Recent literature has shown that early cognitive ability influences the education-health nexus due to the ‘selection effect’. In this paper, we investigate whether midlife cognitive ability has a ‘mediation effect’ in the correlation between educational decisions and midlife outcomes by adopting a structural equation modelling approach. We consider two outcomes, midlife health and midlife earnings. With data from the 1970 British Cohort Study, we construct a measurement model for latent physical and mental health indicators. We also consider an earning variable to explore financial impacts. Our findings suggest that undergraduate and postgraduate education (via midlife cognition) have a positive and significant mediation effect on midlife mental health, while education has a direct negative effect (not significant) on midlife mental health. The two cancelled each other out, leading us to observe no significant overall effect. The effect of undergraduate and postgraduate education on midlife weekly earnings is mainly direct, with a negligible mediation effect. We did not find a significant effect of education on midlife physical health, nor did we find a significant effect of post-compulsory education on midlife outcomes.

The treatment effect of educational decisions on midlife cognitive ability

(with Andrew Jones & Nigel Rice)

Do people with higher early cognitive abilities have a higher probability of moving to the next level of education and which period of early cognitive ability is more important? To analyse these questions, this paper carries out an empirical study to explore the influence of preschool cognitive ability (at age 5) and post-compulsory school cognitive ability (at age 16) on three educational decisions made after compulsory education in Britain, using data from 1970 British Cohort Study. We specify a structural model with a combination of a sequential decision model and a cognitive development model, using confirmatory factor analysis as a measurement model for latent cognitive abilities. A SEM approach is applied for estimation. We find that both early cognitive abilities have positive selection effects on encouraging people to progress to the next stage of their education after compulsory study, although the impact of preschool cognitive ability reduces as the level of education increases. Our findings suggest that post-compulsory school cognitive ability holds a more stable and long-lasting impact on educational decisions relative to preschool cognitive ability.