Understanding predictors of academic achievement has been among the essential research focus of sociologists of education. One of the most influential studies in the field dates to 1960s. James S. Coleman and his colleagues (1966) published their famous report on their scientific attempts to unveil determinants of academic achievement differences of students from different social backgrounds in the US. The results underlined the impact of students’ parental resources and classroom composition based on these resources, more than the impact of schools attended by these students. Despite of providing attention to the impact of individual and cumulative socioeconomic disadvantage of students, the report received critics on the lack of attention to possible school effects. Following the concerning critics, some researchers dedicated their large-scale work more into investigating school effects on academic achievement and ignited an increase in studies for the field of school effectiveness research (Burušić et al., 2016; Scheerens, 2016).

Throughout the years, the field not only yielded a list of significant effects emerging from classroom and school settings, but also highlighted the need for utilizing more precise methodological techniques to account for the nature of nested data while investigating these effects on academic achievement. As an invaluable example, John Hattie (2008)proposed evidences from abundant meta-analyses on how teachers, teaching practices, and schools pose impact on student learning. On the teacher and teaching practices, some of the underlined effects pointed the importance of teaching-occupation-related variables such as expectations from students, teacher training, quality of teaching, and professional development available to teachers, and of strategies related to feedback, learning intentions, and so on (Hattie, 2008). On the school, the importance of school characteristics, school and classroom compositions, school curriculum and classroom effects (Hattie, 2008). While such evidence brought undeniable contributions to the comprehension of multiple interacting effects, Creemers and Kyriakides (2008) introduced a dynamic model of how schools impact academic achievement. Within this model, natural clustering inside formal educational institutions, students within classrooms within schools within educational systems, is illustrated. Due to this natural multilayered structures of schools, the field underlined that data collected from students, classrooms and schools violate the observation independency assumption of traditional analytical techniques (de Leeuw & Meijer, 2008; Hox et al., 2018). Therefore, to model such nested data at multiple levels by controlling for observation dependency, the field recommended the utilization of multilevel modelling as the analytical technique to precisely examine the impacts of these classroom and school factors separately and in combination (Burušić et al., 2016).

Conducting research to probe possible classroom and school effects on academic achievement is especially important in the primary schooling level. The reason behind is ascertain, early in the schooling system, which classroom and school variables significantly explain variations in academic achievement, to be able to diminish the impact of cumulative advantage mechanism, or the Matthew effect (Merton, 1968), resulting in enduring achievement gaps and therefore unequal educational gains and opportunities (Diprete & Eirich, 2006). Besides, results from these scientific attempts potentially aid educational policy makers in developing more effective and efficient approaches to supporting learning, assist school administrations in diagnosing students at risk of lagging behind, and help researchers identify possible research gaps in the field. Consequently, regarding these considerations on the relevance of the classroom and school environment on academic achievement especially in the primary schooling and possible precise contributions from studies utilizing multilevel modelling, the main aim of this systematic review of the existing multilevel modelling studies is to derive and thematize the characteristics of the classroom and school settings impacting academic achievement in primary schooling, by applying strong methodologically-conceptually driven selection criteria.

The research methodology of this study is systematic review. As its research foci, the present review makes a set of inquiries into the following aspects over selected studies: their theoretical and methodological foundations, and significant classroom and school level variables impacting academic achievement in the primary schooling. Accordingly, the selection criteria were for studies to employ multilevel modelling, to examine data collected from primary schooling, and to contain classroom and school levels. After multiple trials, a search query was developed with grouped keywords and controlled terms (wildcards), but without a restriction on publication year or geography. Running the developed query in all fields available for searching on two prominent scientific databases during the last week of February 2021 gathered 112 studies from WoS and 96 studies from ERIC. After removing 25 duplicates, creation of the study pool was finalized with 183 unique study inputs. Before the study screening process, it has been decided to distill studies for significant classroom and school variables separately. Screening for each level were conducted in two stages: title and abstract screening (TAS), and full-text review (FTR). During each stage for each level, two researchers out of three blindly screened the studies in accordance with the selection criteria and the remaining researcher acted as a conflict resolver to reduce the possible researcher bias in screening studies for inclusion. Before conflict resolution, for each step of screening, Cohen’s κ was run to check the agreement level between the screeners. For classroom, the initial yield rate after TAS was 14.75% and the test result revealed a substantial agreement (McHugh, 2012), κ = .649 (95% CI, .498 to .801), SE = .077, p < .01. The final yield rate following FTR was 8.74%, with again a substantial agreement (McHugh, 2012), κ = .743 (95% CI, .471 to 1.000), SE = .139, p < .01. For school, the initial yield rate after TAS was 26.23% and the test result revealed an almost perfect agreement (McHugh, 2012), κ = .874 (95% CI, .794 to .954), SE = .041, p < .01. The final yield rate following FTR was 19.67%, with a substantial agreement (McHugh, 2012), κ = .714 (95% CI, .480 to .948), SE = .119, p < .01. The data extraction was conducted with 16 and 36 studies respectively. Through inductive thematic analysis, the extracted data was examined to categorize the information under similar theoretical, methodological, and contextual/compositional aspects.

The extracted information has been investigated in accordance with the research foci. Regarding the theoretical foundations of the selected studies, many studies focus on school effectiveness research and utilize the systematization of previous empirical findings rather than theoretical considerations. Most conceptual considerations focus on compositional effects on both levels. Very few studies provide linkages to general conceptual frameworks on educational inequalities such as Boudon (1974) and Bourdieu (1986) or Bourdieu & Passeron (1977). The overall reason behind might be the lack of theory particularly in school effectiveness research (Scheerens, 2016). From the methodological foundations, only a few studies focus on earlier grades of primary schooling (e.g., grades levels 1-3), and analyze differences in the respective learning growth of subject specific performances. Regarding outcomes, the focus is often on primary school students’ reading and/or mathematics performances. On the classroom level, the inductive thematic analysis revealed that significant classroom effects are frequently examined regarding student composition, teacher-related determinants, classroom’s socio-physical conditions, and curriculum. On the school level, the analysis revealed that significant school effects are frequently examined regarding student composition, school’s socio-physical conditions, school management, school climate, teacher composition, and curriculum. Explicitly, the results on student composition from both levels indicated that the scholars probed compositional effects from the aspects of socioeconomic background, cognitive and behavioral outputs, ethnicity/language, special education needs, and gender. The brief conclusion from the results underlines that students from advantageous backgrounds, being taught by more occupationally experienced teachers in more prosperous and positive socio-affective learning environments are estimated to have higher academic achievement. Ultimately, more multilevel modelling studies utilizing a longitudinal design, involving earlier grades of primary schooling, and focusing also on other subject-specific performances rather than reading and/or mathematics are needed. Yet, these findings potentially act as a scientific guideline for researchers.

Boudon, R. (1974). Education, Opportunity and Social Inequality: Changing Prospects in Western Society. Wiley. Bourdieu, P. (1986). The forms of capital. In J. G. Richardson (Ed.), Handbook of theory and research for the sociology of education (pp. 241–258). Greenwood Press. Bourdieu, P., & Passeron, J.-C. (1977). Reproduction in education, society and culture (3. pr). Sage. Burušić, J., Babarović, T., & Velić, M. Š. (2016). School Effectiveness: An Overview of Conceptual, Methodological and Empirical Foundations. In N. Alfirević, J. Burušić, J. Pavičić, & R. Relja (Eds.), School Effectiveness and Educational Management: Towards a South-Eastern Europe Research and Public Policy Agenda (pp. 5–26). Springer International Publishing. https://doi.org/10.1007/978-3-319-29880-1_2 Creemers, B. P. M., & Kyriakidēs, L. (2008). The dynamics of educational effectiveness: A contribution to policy, practice and theory in contemporary schools. Routledge. Coleman, J. S., Campbell, E. A., Hobson, C., McPartland, J., Mood, A., Weinfeld, F., & York, R. (1966). Equality of educational opportunity. Washington, DC: U.S. Government Printing. de Leeuw, J., & Meijer, E. (2008). Introduction to Multilevel Analysis. In J. de Leeuw & E. Meijer (Eds.), Handbook of Multilevel Analysis (pp. 1–75). Springer New York. https://doi.org/10.1007/978-0-387-73186-5_1 Diprete, T. A., & Eirich, G. M. (2006). Cumulative advantage as a mechanism for inequality: A review of theoretical and empirical developments. Annual Review of Sociology, 32, 271–297. https://doi.org/10.1146/annurev.soc.32.061604.123127 Hattie, J. (2008). Visible Learning. In Visible Learning: A Synthesis of Over 800 Meta-Analyses Relating to Achievement. Routledge. https://doi.org/10.4324/9780203887332 Hox, J., Moerbeek, M., & van de Schoot, R. (2018). Multilevel Analysis: Techniques and Applications (3, Ed.). Routledge. McHugh, M. L. (2012). Interrater reliability: The kappa statistic. Biochemia Medica, 276–282. https://doi.org/10.11613/BM.2012.031 Merton, R. K. (1968). The Matthew Effect in Science: The reward and communication systems of science are considered. Science, 159(3810), 56–63. https://doi.org/10.1126/science.159.3810.56 Scheerens, J. (2016). Educational Effectiveness and Ineffectiveness. In Educational Effectiveness and Ineffectiveness: A Critical Review of the Knowledge Base. Springer Netherlands. https://doi.org/10.1007/978-94-017-7459-8