Introduction

For the past decades, scholars have debated whether educational tracking increases or decreases overall achievement and whether it enforces or reduces social inequality in achievement. This ongoing debate led to a large body of empirical evidence (Betts, 2011; Hanushek & W ößmann, 2006; Strello et al., 2021; Van de Werfhorst & Mijs, 2010). Overall, this research indicates that tracking increases social inequality without fostering educational achievement (Hanushek & W ößmann, 2006; Matthewes, 2020; Strello et al., 2021; Van de Werfhorst & Mijs, 2010). However, other authors are more skeptical about the generalizability of these findings (Betts, 2011; Esser, 2016) and the debate continues to this day.

Proponents of tracked systems argue that tracking leads to homogenous learning environments (HLE), allowing higher learning rates for all students, resulting in a higher efficiency of tracked systems. However, the proposition of HLE seems to be contradicted by the empirical evidence which suggest that tracked systems do not produce higher average achievement (Hanushek & W ößmann, 2006; Matthewes, 2020; Strello et al., 2021). Although HLE is the central assumption behind tracking, no study has systematically investigated whether this fundamental assumption is actually valid. This study therefore investigates whether tracked education systems lead to homogenous learning environments.

Tracking and Learning Environments

In general, tracking refers to the “practice of assigning students to instructional groups on the basis of ability” (Hallinan, 1994, p. 79) and is found in all education systems in one way or another. In our definition, tracking refers to the existence of early-tracking, where students are sorted into different schools from an age as young as ten.

There is a long and ongoing debate about the effects of (early-)tracked systems (Esser, 2016; Esser & Seuring, 2020; Hallinan, 1994; Oakes, 2005; Van de Werfhorst & Mijs, 2010). Opponents of early-tracking emphasize the possible detrimental effect on social equality (e.g. Esser 2016, p.336). Scholars argue that the educational transition may not strictly follow a meritocratic principle, and that the students’ socioeconomic background may determine the transition in early-tracking systems (e.g., primary and secondary effects, Boudon, 1974). The influence of the family SES is further hypothesized to be stronger at early ages (Hillmert & Jacob, 2010; Mare, 1980; Müller & Karle, 1993). Sorting students at a later age may thus facilitate an accurate assessment of students' scholastic abilities. In a similar fashion, (Brunello & Checchi, 2007, p. 786) argue that observing students' underlying ability at an early age may be difficult and noisy. Thus, early tracking may not lead to homogeneous learning environments. 

Theoretical arguments in favor of tracking, on the other hand, stress the potential positive effects of HLE. Most importantly in this regard are peer effects. Being grouped with others of similar aptitude facilitates learning because students can exert positive influence on each other (e.g. Esser 2016, p.336). Another matter are organizational effects (e.g. Esser 2016, p.336). Teaching and learning might be facilitated, because teachers can tailor their instructions to the specific needs of the homogenous learning group. Tracking students at an earlier age would therefore increase the efficiency of the system, as students are longer exposed to these beneficial conditions.

While both theoretical positions can be appealing, empirical evidence tends to favor the position of opponents of tracking, which suggests that early-tracking increases social inequalities without increasing overall achievement levels (Hanushek & W ößmann, 2006; Matthewes, 2020; Strello et al., 2021). We believe this inconsistency needs further explanation and investigate the fundamental assumption behind tracked systems: the HLE assumption. If tracked systems do not produce HLE this could hint why we observe the described empirical patterns

Data and Methods To test our theoretical assumptions, we draw on several waves from the education studies PIRLS, TIMSS 4th grade, TIMSS 8th grade, and PISA. The homogeneity of the learning environment, our dependent variable, was measured using the intraclass correlation coefficient (ICC) of scholastic achievement within schools. The ICC indicates how similar students within a school are to each other with respect to achievement. For our treatment variable, we differentiate between early- and late-tracking education systems. Early- and late-tracking countries were identified using the grade of the first tracking. Education systems with the first tracking before the 8th grade respectively were categorized as early-tracking systems. Education systems with no tracking or with the first tracking after the 8th grade are considered late-tracking countries. We followed and reviewed the coding of Strello and colleagues (2021) and supplemented the remaining countries with the information provided by TIMSS Wiki and Eurydice. In order to identify the effect of tracking on the homogeneity of the learning environment, we used a Difference-In-Differences (DiD) approach (Wing et al., 2018). The main advantage of the DID approach is that effects are estimated only by using change within units of interest, in our case, countries (Jakubowski, 2010). Thus, we can investigate differences in the homogeneity of the learning environment between early-tracking and late-tracking education systems by comparing differences between the education systems before (among 4th-grade students) and after the tracking took place (among 8th-grade students). Lastly, we will check the robustness of our results using a number of different HLE operationalizations (e.g. dissimilarity index or Gini coefficient) samples (e.g. OECD countries only) and operationalizations of early-tracking.

Expected Results Our paper attempts to test a fundamental proposition associated with tracked education systems: the creation of homogenous learning environments through tracking. In this regard, we are agnostic towards any result but stress the two opposing positions. That is, proponents of early-tracked systems would expect considerably higher homogeneity in tracked systems -as compared to integrated systems- since the early selection, sorts students into different schools according to (perceived) ability. Opponents of tracked systems, on the other hand, would expect that the early sorting process is significantly confounded by “noise”, either because true “ability” cannot perfectly be observed at this age and/or because it is confounded by factors of the family background. This would result in a much weaker association between tracked systems and homogenous learning environments.

Betts, J. R. (2011). The economics of tracking in education. In Handbook of the Economics of Education (Vol. 3, pp. 341–381). Elsevier. Boudon, R. (1974). Education, opportunity, and social inequality: Changing prospects in western society. Wiley. Brunello, G., & Checchi, D. (2007). Does school tracking affect equality of opportunity? New international evidence. Economic Policy, 22(52), 782–861. Esser, H. (2016). Bildungssysteme und ethnische Bildungsungleichheiten. In C. Diehl, C. Hunkler, & C. Kristen (Eds.), Ethnische Ungleichheiten im Bildungsverlauf (pp. 331–396). Springer Fachmedien Wiesbaden. Esser, H., & Seuring, J. (2020). Kognitive Homogenisierung, schulische Leistungen und soziale Bildungsungleichheit: Theoretische Modellierung und empirische Analyse der Effekte einer strikten Differenzierung nach den kognitiven Fähigkeiten auf die Leistungen in der Sekundarstufe und den Einfluss der sozialen Herkunft in den deutschen Bundesländern mit den Daten der „National Educational Panel Study“ (NEPS). Zeitschrift für Soziologie, 49(5–6), 277–301. https://doi.org/10.1515/zfsoz-2020-0025 Hallinan, M. T. (1994). Tracking: From theory to practice. Sociology of Education, 67(2), 79–84. Hanushek, E. A., & W ößmann, L. (2006). Does Educational Tracking Affect Performance and Inequality? Differences‐ in‐Differences Evidence Across Countries. The Economic Journal, 116, C63–C76. https://doi.org/10.1111/j.1468-0297.2006.01076.x Hillmert, S., & Jacob, M. (2010). Selections and social selectivity on the academic track: A life-course analysis of educational attainment in Germany. Research in Social Stratification and Mobility, 28(1), 59–76. Jakubowski, M. (2010). Institutional tracking and achievement growth: Exploring difference-in-differences approach to PIRLS, TIMSS, and PISA data. In Quality and inequality of education (pp. 41–81). Springer. Mare, R. D. (1980). Social background and school continuation decisions. Journal of the American Statistical Association, 75(370), 295–305. Matthewes, S. H. (2020). Better Together? Heterogeneous Effects of Tracking on Student Achievement. CEP Discussion Paper No. 1706. Centre for Economic Performance. Müller, W., & Karle, W. (1993). Social selection in educational systems in Europe. European Sociological Review, 9(1), 1–23. Oakes, J. (2005). Keeping track: How schools structure inequality. Yale University Press. Strello, A., Strietholt, R., Steinmann, I., & Siepmann, C. (2021). Early tracking and different types of inequalities in achievement: Difference-in-differences evidence from 20 years of large-scale assessments. Educational Assessment, Evaluation and Accountability, 33(1), 139–167. https://doi.org/10.1007/s11092-020-09346-4 Van de Werfhorst, H. G., & Mijs, J. J. (2010). Achievement inequality and the institutional structure of educational systems: A comparative perspective. Annual Review of Sociology, 36, 407–428. https://doi.org/10.1146/annurev.soc.012809.102538 Wing, C., Simon, K., & Bello-Gomez, R. A. (2018). Designing difference in difference studies: Best practices for public health policy research. Annual Review of Public Health, 39. https://doi.org/10.1146/annurev-publhealth-040617-013507