The study's overall purpose is to explore the formation of student academic self-concept (ASC) in the subjects of English and mathematics. ASC is commonly defined as self-perceived academic ability and is related to cognitive and non-cognitive outcomes such as academic engagement, goal-setting, task choice, persistence and effort, intrinsic motivation, strategy use, academic achievement, and future career selection (Bong & Skaalvik, 2003; Marsh et al., 2019). When students perceive their previous experiences of academic activities to be positive and when they perceive that they are capable of managing future academic activities, it is thus an advantage that goes beyond immediate academic success. Rather, ASC has been shown to have prolonged effects (Marsh et al., 2001).

   Because ASC frequently has been shown to be important for student success, much research has been dedicated to explaining how it is formed. The main explanation is the big-fish-little-pond effect (BFLPE), which posits that equally abled students perceive their abilities differently depending on their context (Marsh et al., 2008). A student in a high-achieving context would rate their ability to be lower than a student in a lower-achieving context, even if both students have the same abilities.

   In 1962, tracking was introduced in the subjects of English and mathematics in all secondary schools in Sweden (Grades 7-9). With recommendations from teachers, students were to choose between advanced and general courses in the two subjects (Marklund, 1985). The general courses were easier and given at a slower pace than the advanced courses and tended to have lower class-average achievement. Tracking is no longer a formal practice in Swedish compulsory education, but it commonly occurs when teachers organise education in Sweden and internationally (Trautwein et al., 2006).

   ASC is a well-researched area, but so far, only a few studies have conducted longitudinal analyses to investigate effects over time. There is also a need for studies that look at how ASC is affected by school systems with some form of tracking (i.e., ability stratification, ability grouping etc.). The specific purpose of the study is to explore the effects of non-tracking and tracking in secondary school on ASC in upper secondary schools. With longitudinal data from the 1980s and 90s, ASC will be measured in Grade 6 (pre-tracking) and Grade 10 (post-tracking).

Previous Research

   In a longitudinal study, Marsh et al. (2001) compared students from former East and West Germany (N = 2 778). They found that when East and West Germany reunited and the schools merged, the students who had attended the selective and ability-stratified schools in West Germany were more strongly affected by the negative BFLPE when compared to the East German students. Before the reunification, East German students had not experienced an ability-stratified school system. The difference between the merged students decreased with time when the former East German students became integrated with the more selective school system. Overall, the findings of Marsh et al. (2001) indicate that school policies and systems may have an impact on the formation of student ASC.

   Similarly, Liem et al. (2013) found that compulsory school students in low-ability streams in English and mathematics had higher self-concepts than students in high-ability streams when student achievement was controlled for. However, Herrmann et al. (2016) investigated the German within-school track system (N = 1 330) and found that the negative BFLPE for students in the advanced mathematics track disappeared when they controlled for positive assimilation effects. The positive assimilation effect is similar to the basking-in-reflected-glory (BIRG) effect, which both refer to the notion that attending a high-achieving class or school positively affects ASC.

Participants and procedure Data will be retrieved from the Swedish longitudinal project Evaluation through Follow-up (UGU), compiled by Statistics Sweden (Härnqvist, 2000). The sampling was a two-step stratified procedure, where municipalities were selected in the first step and classes in the second step. The UGU samples are nationally representative of their respective populations. Four birth cohorts will be used in the study, 1967 (N = 9 104), 1972 (N = 9 498), 1977 (N = 4293), and 1982 (N = 8 805). Cohorts 1967, 1972, and 1977 experienced tracking and will be merged to get a bigger sample. Cohort 1982 did not experience ability-streamed courses and will function as a control group. UGU consists of register, survey, and test data. Survey data was first collected in Grade 6 and then for a second time in upper secondary school. For cohorts 1967, 1972, and 1977 the second data collection occurred in Grade 10 and for cohort 1982 it occurred in Grade 12. Survey data from Grades 6 and 10/12 will be used to measure ASC pre- and post-tracking. To deal with missing data, calibration weights and full information maximum likelihood (FIML) estimation will be used to correct for bias due to non-participation. Measures and variables Cohorts 1967, 1972, and 1977 answered identical questions in Grade 10, while cohort 1982 answered similar but not identical questions as the other three cohorts. Measures of ASC will be constructed to be as similar as possible between the three earlier cohorts and cohort 1982. Factors will be created with indicators of students’ ASC, for example, “What kind of arithmetic skills do you think you have?” and “Did you experience any problems with arithmetic in secondary school”. Achievement will be operationalized by grade point average (GPA) from Grade 9 and by cognitive ability from Grade 6. Cognitive ability will be measured with three tests measuring students’ verbal, spatial, and inductive abilities. Gender and parental education will also be included. Method of Analysis First, descriptive analyses will be calculated. Measurement models will then be constructed in Mplus with confirmatory factor analysis (CFA), to create latent variables for ASC in Grades 6 and 10/12. Lastly, longitudinal structural equation modelling (LSEM) will be used. The tracking system enables a quasi-experimental research design, that in turn makes it possible to investigate the effect of tracking on subsequent ASC with LSEM and the control group that did not experience tracking.

Regarding the possible outcomes of the study, two contradictory effects are relevant to consider. It concerns the previously mentioned BFLPE as well as the basking-in-reflected-glory (BIRG) effect. The BIRG effect predicts that when students perceive their school or class (i.e., their reference group) to have high status, it affects their self-concepts positively (Marsh et al., 2000). The glory of attending a high-status group thus reflects on the individuals in the group, regardless of individual achievement level. In contrast, the BFLPE predicts that attending a high-achieving group affects students’ self-concept negatively, because of negative social comparison processes. Even if both effects concern the formation of self-concept, research has indicated that the BFLPE is the most dominant effect of the two (Marsh et al., 2000). I.e., the negative social comparison effect tends to have a greater impact on students’ self-concept than the positive effect of attending a high-status group. In the present study, the BFLPE hypothesis would be that students who attended the advanced courses in English and mathematics reported lower ASC in Grade 10 because their ASCs were negatively affected by the comparisons with high-ability peers in secondary school. However, it may also be that the BIRG effect is present rather than the BFLPE, which would mean that students in the advanced courses express higher ASC due to reflected glory.

Bong, M., & Skaalvik, E. M. (2003). Academic Self-Concept and Self-Efficacy: How Different Are They Really? Educational Psychology Review, 15(1), 1–40. Herrmann, J., Schmidt, I., Kessels, U., & Preckel, F. (2016). Big fish in big ponds: Contrast and assimilation effects on math and verbal self‐concepts of students in within‐school gifted tracks. British Journal of Educational Psychology, 86(2), 222– 240. Härnqvist, K. (2000). Evaluation through follow-up. A longitudinal program for studying education and career development. In C.-G. Janson (Ed.), Seven Swedish longitudinal studies in behavioral science (pp. 76–114). Stockholm: Forskningsrådsnämnden. Liem, G. A. D., Marsh, H. W., Martin, A. J., McInerney, D. M., & Yeung, A. S. (2013). The Big-Fish-Little-Pond Effect and a National Policy of Within-School Ability Streaming: Alternative Frames of Reference. American Educational Research Journal, 50(2), 326–370. Marklund, S. (1985). Skolsverige 1950-1975 D. 4 Differentieringsfrågan. Stockholm: Liber Utbildningsförlaget. Marsh, H. W., Köller, O., & Baumert, J. (2001). Reunification of East and West German School Systems: Longitudinal Multilevel Modeling Study of the Big-Fish- Little-Pond Effect on Academic Self-Concept. American Educational Research Journal, 38(2), 321–350. Marsh, H. W., Kong, C., & Hau, K. (2000). Longitudinal multilevel models of the big-fish-little-pond effect on academic self-concept: Counterbalancing contrast and reflected-glory effects in Hong Kong schools. Journal of Personality and Social Psychology, 78(2), 337–349. Marsh, H. W., Pekrun, R., Parker, P. D., Murayama, K., Guo, J., Dicke, T., & Arens, A. K. (2019). The murky distinction between self-concept and self-efficacy: Beware of lurking jingle-jangle fallacies. Journal of Educational Psychology, 111(2), 331–353. Marsh, H. W., Seaton, M., Trautwein, U., Lüdtke, O., Hau, K. T., O’Mara, A. J., & Craven, R. G. (2008). The Big-fish–little-pond-effect Stands Up to Critical Scrutiny: Implications for Theory, Methodology, and Future Research. Educational Psychology Review, 20(3), 319–350. Trautwein, U., Lüdtke, O., Marsh, H. W., Köller, O., & Baumert, J. (2006). Tracking, Grading, and Student Motivation: Using Group Composition and Status to Predict Self-Concept and Interest in Ninth-Grade Mathematics. Journal of Educational Psychology, 98(4), 788 – 806.