Although the effect of tracking on educational inequality has long been the subject of education research (see Schmidt & Burroughs 2012 for a summary), until recently most of that work has relied on fairly blunt measures of curricular inequality (e.g. earliest age of tracking) and emphasized between-school differences. Recent research making use of student-level indicators of opportunity to learn (OTL) suggests widespread within-school inequality that is systematically related to student socioeconomic status (Schmidt et al. 2015). Schmidt et al. found that roughly a third of the relationship between SES and student mathematics scores on the 2012 PISA was due to the association of SES to OTL. In all OECD countries there was a statistically significant relationship between OTL and student socioeconomic status. Further, the study uncovered considerable variation in which inequalities in SES-based inequalities in OTL were attributable to between or within school differences.

The Schmidt et al. study’s focus on OECD countries, which are generally wealthy and confined to particular geographic regions, raises several important questions. First, do these results hold for non-OECD countries, many of whom (although not all) are developing countries? As policymakers in the global south work to improve educational opportunities and average academic performance as part of an economic development strategy, it is important to consider the degree to which their generally higher levels of background socioeconomic inequality are being either reinforced or mitigated within the educational system. A subsidiary question is whether these relationships hold for former Communist countries, which have been shown to have a distinct pattern of educational inequality and achievement (Jerrim).

Second, there are potential concerns with the studies’ reliance on PISA’s index of social, economic and cultural status (ESCS). Rutkowski and Rutkowski (2012) have argued that the PISA index has weaker reliability in many low-income countries, which may complicate efforts to relate SES to OTL using the ESCS indicator.

Third, the Schmidt et al. study models the size of SES-related gaps using pooled within-country SES quartiles, taking the wealthiest 25% and poorest 25% of students in the entire country to establish top and bottom thresholds, and then simply taking the difference in mean OTL and PISA math scores between the top and bottom quarter of students. This approach results in the exclusion of schools without members in both quartiles. Given the apparent importance of within-school inequalities, the obvious question is whether these SES “gaps” persist when inequalities are defined within each school rather than within each country – a strategy made possible by the interval-level character of the ESCS index.

In the proposed paper we test the robustness of the Schmidt et al. findings by addressing the following questions: 1) To what extent do inequalities in OTL exacerbate SES inequalities in non-OECD countries, as they have been shown to do in OECD nations? 2) Do measures of SES other than the ESCS index also yield consistent associations of SES to OTL? 3) Are SES-related inequalities in OTL and student achievement to be found when the cutoffs for high and low SES are allowed to vary by school? 

Hirsch, F. (1978). Social Limits to Growth. Routledge: London, England. Jerrim, J. and Macmillan, L. (2014). “Income inequality, intergenerational mobility and the Great Gatsby Curve: is education the key?” Social Forces 94 (2): 505-533. Rutkowski, D. and Rutkowski, R. (2013). “Measuring Socioeconomic Background in PISA: One Size Might not Fit all.” Research in Comparative and International Education 8 (3): 259-278. Schmidt, W. H. and Burroughs, N.A. (2013) "Opening the Black Box: Prospects for Using ILSAs to Explore Classroom Effects." Research in Comparative and International Education 8 (3): 236-247. Schmidt, W.H., Burroughs, N.A., Zoido, P., and Houang, R. (2015) “The Role of Schooling in Perpetuating Educational Inequality An International Perspective.” Education Researcher 44 (7): 371-386.