How do the environments that children face interact with their skills to determine school success?

The attainment gap between disadvantaged pupils and their more advantaged peers at the end of secondary school is growing (EPI, 2020). Globally, this is not unusual, and the persistence of such inequities despite substantial academic interventions by schools is troubling. In response, policy-makers and academics have begun to look further afield for solutions. The recent professional and academic focus on non-cognitive skills is one such direction, beginning to illuminate the importance of pupils' social and emotional skills for their academic and lifelong success. Another is the increased scrutiny of potential biases in teacher grades, recommendations, and mentoring. However, it is both surprising and noteworthy that very little focus has yet been placed on the interactions between these two areas of enquiry: namely, how differences in the environment that pupils find themselves in interact with their own characteristics and behaviours to produce outcomes.

The importance of these recent research findings to the education sector cannot be over-stated. In the UK, discussions around the attainment gap remains focused on cognitive and noncognitive skills deficits of disadvantaged children, and on interventions that are intended to 'close' this skills gap (e.g. EEF, 2020). As a teacher, I have attempted to implement a number of these interventions myself and have seen the unrewarding results that come from trying to 'change' disadvantaged pupils to fit the requirements of the school, rather than the support of the school change to fit its most vulnerable pupils. Instead, this literature begins to answer the crucial question of whether the skills and environments that are lauded by many school leaders and policy makers - such as conscientiousness, academic foci on maths and English, and behaviour policies that disincentivise risk taking - actually exacerbate existing inequities by rewarding and supporting skills that only benefit more-advantaged pupils. In this way, it suggests that we must know more about the behaviour and capabilities that actually foster success for disadvantaged children and design interventions, and schools, suitably.

Nonetheless more research needs to be done to better understand the how the environments that children face interact with their skills to determine school success. In particular, as the returns to noncognitive and cognitive skills for different groups is likely to differ by context, it is important to determine whether the associations that are apparent in Lundberg's research also hold true in the UK. In addition, Lundberg's paper quantifies academic attainment primarily as a measure of persistence, using high school completion, college enrolment and college completion as indicators of success. While these are important measures, using examination grades as a dependent variable - when controlling for prior academic attainment - may produce very different results. Lastly, although Lundberg convincingly make the case for causality between personality traits and outcomes using correlational results, more robust experimental or quasi-experimental research designs which estimate whether or not such a causal relationship is truly extant are needed.

The present research therefore seeks to answer the question of whether traits that are conducive to educational achievement differ systematically across family and school - and possibly social - contexts. I will do this by firstly examining the association of each of the 'Big Five' personality traits with educational attainment and labour market outcomes, controlling for covariates including prior educational attainment and allowing results to differ by gender and socioeconomic status. Research and results permitting, I will then examine potential causality in the relationship between measured behaviours indicative of certain personality traits and outcomes, again differentiating by gender and socioeconomic status.