Association for Public Policy Analysis & Management

A growing literature has found that non-cognitive factors, such as an individual’s motivation, eagerness to succeed, agreeableness, or ambition, affect human capital accumulation (Cunha, Heckman, & Schennach 2010). In the education context, non-cognitive skills affect grades and test scores by affecting how much attention students pay to instruction from their teachers, how organized they are in doing homework and preparing for exams, whether they get disciplined for inappropriate behavior at school, and in various other ways. More specifically, recent research has found that individuals’ item nonresponse in survey questionnaires, as a measure of non-cognitive skills, is also related with their performance in school (Borghans & Schils, 2012; Zamarro, Hitt & Mendez, 2016).

Still, there are few intersectional studies that explore the gender and socioeconomic status differences in non-cognitive skills as an explanation for academic achievement (Becker, Hubbard & Murphy, 2010). Despite the issue of the disparity in educational attainment between males and females been driving public policy debates in recent years, there is little research on the differences of non-cognitive skills by gender and by socioeconomic status. In this paper, we address this gap in the literature to extend the analysis of emerging gender and socioeconomic differences in academic achievement by exploring the extent to which these differences are driven by non-cognitive skills, such as conscientiousness and persistence. Based on previous research that provides compelling evidence that measures of effort capture student non-cognitive skills (Borghans & Schils, 2012), we will use Zamarro, Hitt and Mendez’s (2016) measures of student effort as proxies for non-cognitive skills, namely student effort, as measured by non-response rate, careless answering and test decline.

Thus, this study aims to explore the differences of non-cognitive skills, proxied by student effort as measured by non-response rate, careless answering and test decline, by gender, across social classes, in three countries: Japan, U.S., and Thailand. The results of this study would inform education policy by analyzing the influence of gender and socioeconomic status differences on non-cognitive skills, proxied by measures of students’ effort developed by Zamarro, Hitt and Mendez (2016). Through the application of intersectionality framework, differences among multiple grounds of identity can be better acknowledged in the policymaking process.