Large-scale Assessments in Education (Apr 2018)
Personality, cognitive skills and life outcomes: evidence from the Polish follow-up study to PIAAC
Abstract
Abstract Background There is a growing literature providing evidence on the importance of non-cognitive skills for life outcomes. However, to date there is limited evidence on the gains from incorporating such measures into large-scale competence surveys. Methods We investigate the relationship between personality traits and eight important life outcomes: educational attainment, labour market participation, employability, wages, job satisfaction, health, trust and life satisfaction measured in the Polish follow-up study to PIAAC. The study assesses two short scales: the Big Five Inventory and Grit. First, we compare explanatory power of personality traits to that of cognitive skills measured by PIAAC. Second, an incremental validity of Grit after controlling for the Big Five dimensions is assessed. Results The analyses show that differences in personality traits are important in explaining differences in life outcomes. Educational attainment is more strongly related to cognitive skills, while for wages, the explanatory power of personality and cognitive skills is similar. For most of the subjective outcomes, the Big Five traits outperform cognitive skills in predictive power. Conscientiousness is positively related to most of the outcomes analysed while Neuroticism has a negative relationship. After controlling for sociodemographic characteristics and cognitive skills, Big Five traits add explanatory power to all models except for employability. Grit explains some additional variation in educational attainment and in a number of subjective outcomes: health, trust, job and life satisfaction, even after adjusting for the effects of cognitive skills and Big Five traits. Conclusions Given the potential benefits and relatively small burden on respondents in terms of required time it seems advisable to incorporate measures of personality traits into competence surveys as they contribute to explaining the variability in policy-relevant outcomes. The use of the Big Five Inventory seems preferable to Grit when a broad range of life outcomes is of interest, as the former covers multiple aspects of personality. However, using both scales offers an improvement in explanatory power.
Keywords
