centre for microdata methods and practice

ESRC centre

cemmap is an ESRC research centre

ESRC

Keep in touch

Subscribe to cemmap news

The impact of early cognitive and non-cognitive skills on later outcomes

Authors: Pedro Carneiro , Claire Crawford and Alissa Goodman
Date: 01 January 2007
Type: External publications,

Abstract

When describing the determinants of economic or social outcomes, economists often have a very simplified view of skill. Failure to take into account the fact that skill is intrinsically a multidimensional object may misguide both research and the design of social policy.

In this paper, we analyse the consequences and determinants of cognitive and noncognitive (social) skills at age 11, using data for Great Britain from the National Child Development Survey (NCDS). We document the importance of these skills for schooling attainment, labour market outcomes and social behaviours at various ages, and analyse the role of family background and the home learning environment in the formation of these skills.

We find that an overall measure of non-cognitive skill is important for a host of outcomes, including whether or not an individual stays on at school beyond the age of 16, whether they have obtained a degree by age 42, employment status at age 42, work experience between ages 23 and 42, wages at age 42, smoking at age 16, truancy before age 16, exclusion from school, teenage pregnancy, involvement with crime (ages 16 and 42), and health at age 42. Furthermore, the impact of this measure of non-cognitive skill does not differ in any systematic way across particular subgroups of interest (including those defined according to parental education, or father's socioeconomic status).

We go on to split this measure of non-cognitive skill into twelve different domains. As an example, we find that 'inconsequential behaviour' at age 11 (for example, misbehaviour in class) is associated with a reduction in the probability that an individual will stay on at school beyond age 16, a reduction in their wages at age 42, an increase in the likelihood that they will be a heavy smoker at age 16, and an increase in the probability that they will have played truant or been involved in crime by age 16. Further, we find that depression at age 11 is associated with a reduction in the probability that an individual will have obtained a degree by age 42, an increase in the probability that the individual will be a heavy smoker at age 16, an increase in the likelihood that they will have been excluded from school, and an increase in the probability that they will report symptoms of depression at age 42.

These findings together make it clear that a vision of the world in which skill is thought of as a one-dimensional object is extremely inadequate.

Further, we show that both cognitive and non-cognitive skills are strongly dependent on family background and other characteristics of the home learning environment, and that this is likely to be for both genetic and environmental reasons. More importantly, our work suggests that social skills may be more malleable than cognitive skills, which - if true - suggests that there may be greater scope for education policy to affect social skills rather than cognitive skills.

Download full version

Publications feeds

Subscribe to cemmap working papers via RSS

Search cemmap

Search by title, topic or name.

Contact cemmap

Centre for Microdata Methods and Practice

How to find us

Tel: +44 (0)20 7291 4800

E-mail us