Overview
Studies have documented that social and emotional skills are increasingly important in the workplace and attract higher wages. Socio-emotional skills such as personality traits, goals, character, motivations, and preferences are successful in predicting educational attainment, labour market success, health, and criminality. There is a debate in the literature on how best to measure social skills, including using performance on tasks, self-reported questionnaires, and observed behaviours (Kautz et al., 2014). Several papers have shown that measuring socio-emotional (non-cognitive) skills by behaviours observed in the adolescent years is a promising approach that is free of reference bias. These behaviours can include risky and reckless behaviours such as stealing from a store, purposefully damaging property or skipping class as well as positive behaviours such as taking part in extra-curricular activities (Kautz et al., 2014; Heckman et al., 2014).
We are advertising a PhD studentship to recruit a student to build on existing research by investigating how social skills are produced, how they relate to observed outcomes and how they mediate other behaviours and outcomes. Throughout, the focus will be on behavioural measures of social skills. A central data source will be the UK Longitudinal Household Survey (UKHLS) which contains a Youth Panel administered to 10-15 year olds. It contains longitudinal individual data on measures of self-esteem and anxiety (the Strength and Difficulties Questionnaire), bullying, truancy, smoking and alcohol use and other risky behaviours, vandalism, happiness (mental health) and social media use for a representative sample of adolescents and young adults. The data are linked to administrative school records including individual-level test results and school-level characteristics. Linking siblings in the data and comparing outcomes for siblings within the same family allows the researcher to control for unobserved differences between families.
One suggested avenue of enquiry for this studentship is to investigate the socio-economic status (SES) gradient in the gender gap in education. One of the most dramatic changes in the landscape of education in the UK has been that girls have overtaken boys. Today, girls are out-performing boys in school, are 35% more likely to go to university, and do better once they get there. But an important, under-studied fact is that the gender (i.e. the underperformance of boys) gap in educational attainment is larger, and perhaps increasing, among families of low socioeconomic status. Recent research using US data suggests that an important reason that boys perform less well than girls in low socio-economic status (SES) households is that they have more behavioural problems. They trace this to being more closely linked to family circumstances (absent fathers, less educated mothers) than to the quality of schools and neighbourhoods (Autor et al. 2017). The PhD will describe the attainment gradient and its variation with age, and investigate behaviours that may play a causal role.
Another possible line of research includes using the richness of the UKHLS data to explore the predictive power of various types of behaviours for outcomes including exam results, labour market success, health and family formation. Further, the PhD may investigate how social and cognitive skills interact with each other and inputs from schools and parents to produce outcomes in the labour market.
References
Autor, David, David Figlio, Jeffrey Roth and Melanie Wasserman (2017): Family disadvantage and the gender gap in behavioral and educational outcomes [R&R AEJ: Applied]
Kautz et al. (2014), Fostering and Measuring Skills: Improving Cognitive and Non-Cognitive Skills to Promote Lifetime Success, National Bureau of Economic Research Working Paper Series No. 20749
Heckman, J. J., S. H. Moon, and R. Pinto (2014). The effects of early intervention on abilities and social outcomes: Evidence from the Carolina Abecedarian study. Unpublished manuscript, University of Chicago.