Family Inequality, School Inequalities, and Mathematics Achievement in 65 Countries: Microeconomic Mechanisms of Rent Seeking and Diminishing Marginal Returns

Abstract

Background While many studies show that greater economic inequality widens the achievement gap between rich and poor students, recent studies indicate that countries with greater economic inequality have lower overall student achievement. Purpose This study explores whether family inequalities (family income) or school inequalities (educational materials or teachers with university degrees) reduce overall student achievement through micro-economic mechanisms, such as fewer educational resources (via rent-seeking) or inefficient resource allocation (via diminishing marginal returns). Population/Participants/Subjects The Organisation for Economic Cooperation and Development's Programme for International Student Assessment (OECD-PISA) selected 475,760 representative fifteen-year-olds and their principals from 18,094 schools in 65 countries. Research Design In this secondary analysis, we tested whether family or school inequalities were related to students’ mathematics test scores, and whether fewer educational resources or inefficient resources allocation mediated these relationships. Data Collection and Analysis Each student received a mathematics test. The students and their principals also received a questionnaire. World Bank economic data on each countries were merged with the OECD-PISA data. To analyze this data, we used item response models, Warm indices and multilevel analyses. Findings/Results In countries with greater family inequality (GDP Gini) or school inequalities (of educational materials or teacher quality), students had lower mathematics achievement. The results were similar in all student subsamples (high vs. low SES; high vs. low achievement). As the mediation results for each inequality differed, they suggest that these inequalities operate through different mechanisms. Family inequality and school inequality of teacher quality are linked to fewer teachers with post-secondary education and lower mathematics achievement. Meanwhile, school inequality of educational resources is linked to diminishing marginal returns and lower mathematics achievement. Conclusions/Recommendations Family inequality and school inequalities (educational materials, teacher quality) are distinct inequalities that are all linked to lower mathematics achievement, but not substantially correlated with one another. Thus, each inequality can be addressed separately. As none of the subgroups of students (not even the richest ones) benefit from any of the inequalities, disseminating the results widely can help more laypeople (especially the richest ones) recognize their mutual benefit in reducing these inequalities –or reduce their inclination to support policies that exacerbate these inequalities. As reducing family inequality can be extremely costly and politically controversial, a strategic intervention at the inequality mechanism level (e.g., increasing teacher quality in schools with few high quality teachers) might be improve mathematics achievement more effectively.

Streaming, tracking and reading achievement: A multilevel analysis of students in 40 countries.

“…These results suggest that classmates shared educational materials, ideas, and experiences with a student to aid his or her reading achievement (e.g., Kerr, Pekkarinen, & Uusitalo, 2013). As classmate family SES and national income both show diminishing marginal returns on reading achievement, they benefit lower SES students more than higher SES students, consistent with earlier studies (e.g., Chiu, 2015). Furthermore, it suggests that mixing high SES students with low SES students in a classroom (diversity) would increase the benefits of classmate SES (Chiu, 2015), which is supported by our classmate family SES variance results; when This document is copyrighted by the American Psychological Association or one of its allied publishers.…”

“…As classmate family SES and national income both show diminishing marginal returns on reading achievement, they benefit lower SES students more than higher SES students, consistent with earlier studies (e.g., Chiu, 2015). Furthermore, it suggests that mixing high SES students with low SES students in a classroom (diversity) would increase the benefits of classmate SES (Chiu, 2015), which is supported by our classmate family SES variance results; when This document is copyrighted by the American Psychological Association or one of its allied publishers.…”

“…Classmate family variables were also linked to students’ reading achievement (Table 2, Model 4). If classmate family SES exceeded the mean by 10%, students averaged 4 points higher in reading achievement, and this nonlinear relation was smaller for higher classmate family SES ( diminishing marginal returns , Chiu, 2015; see Appendix C), supporting the classmate sharing hypothesis H-7. Moreover, students with classmates who had 10% more education resources at home averaged 5 points higher reading achievement, and this technically nonlinear relation was nearly linear (see Appendix C).…”

The Impacts of School Counselor Ratios on Student Outcomes: A Multistate Study

“…Because students in poorer families or poorer schools underperform other students with more family or school resources (Chiu, 2015; Chiu et al, 2017), educators advocate for greater equity in school funding to narrow achievement gaps (Education Trust, 2018). Among students from low-income backgrounds, those whose schools received 10% more school funding than others across all 12 years of public school completed 0.46 additional years of education, earned 9.6% more income, and were 6.1% less likely to be in poverty as an adult (Jackson et al, 2016, p. 160).…”

Inadequacies in the SES–Achievement model: Evidence from PISA and other studies

“…The OECD (2019b, p. 44) favours comprehensive systems asserting that educational differentiation exacerbates SES inequalities. Other aspects of schools postulated as important to SES inequalities in achievement include school quality (Rouse & Barrow, 2006), teacher quality (Chiu, 2015), school effectiveness (Hobbs, 2016), school climate (Berkowitz et al, 2017) and school resources (Greenwald et al, 1996; but see Hanushek, 1997).…”