Time: Friday, 30/Aug/2024: 11:30 - 13:00 Session Chair: Joe O'Hara |
Location: Room 013 in ΧΩΔ 02 (Common Teaching Facilities [CTF02]) [Ground Floor] Cap: 60 |
Paper
Instructional Practice, Teacher Characteristics and Their Influence on Student Achievement in Science: A Study of TIMSS 2019 in Sweden
University of Gothenburg, Sweden
Presenting Author:It has long been acknowledged that science skills play a crucial role in fostering economic development (Hanushek & Woessmann, 2012) and technological innovation (Varsakelis, 2006). Therefore, governments around the world are searching for ways to effectively enhance science education. Teachers and their instructional quality play an important role in student achievement and learning (Harris & Sass, 2011). It is also emphasized that teachers are one of the essential factors to enhance student skills and knowledge improvement (Harris & Sass, 2011). However, among several factors associated with students, teaching strategies, school, and home, teacher quality has an important role in student achievement.
The teacher quality framework (Goe, 2007) suggests that a teacher’s college degrees, certificates, and their test scores, among a group of inputs, can indicate who might be a successful teacher inside a classroom. However, teaching quality is not only defined by teacher certification and training but is also explained by what teachers do inside a classroom and how they teach, i.e., their classroom practices. Blömeke, Olsen and Suhl (2016) evaluated the relationship between educational input and process properties of schooling, and students’ cognitive outcomes with TIMSS 2011 data. The study revealed that teacher quality was significantly related to instructional quality and student outcome, while instructional quality was not a good predictor of student outcome.
Teacher’s knowledge of subject-matter, teaching skills, personal characteristics, and professional development have been found to be among the most effective characteristics of teachers (e.g., Toraman, 2019). However, no consensus on the essential teacher qualifications that explain students’ academic performance has been reached (Scheerens & Blömeke, 2016; Lee & Lee, 2020). Moreover, despite the emphasis on improving teaching qualifications (Goe, 2007), students from socioeconomically disadvantaged households and ethnic minorities are less likely to receive instruction from qualified teachers, since less qualified teachers are concentrated in schools and classrooms teaching students with low socioeconomic status and academic achievement (Luschei & Jeong, 2018).
Additionally, educational equity and quality are considered central points of Swedish school policy (Kelly et al., 2020). The Swedish Education Act underscores the school system’s mission of offering equal education quality, learning opportunities, and support to all students regardless of their background characteristics, and the type of schools they are attending (Swedish Education Act, 2010).
Against this background, the main objective of this study is to investigate the relationship between teachers’ experience, education, and their instructional practice, while controlling for students’ socioeconomic background and classroom SES composition. The study also aims to examine the schools’ compensatory effect of the teacher -related factors for educational equity, which is measured as the influence of home educational resources on students’ science achievements. According to the national curriculum for the compulsory school in Sweden, science is separated into the subjects of biology, chemistry, and physics. Therefore, the following research questions will be scrutinized in each of the science subjects:
1- Are teachers’ instructional practice, teachers’ experience and education, and classroom SES composition significantly related to student science achievement, controlling for students’ family background?
2- How do teachers’ experience and education, and classroom SES composition relate to their instructional practice?
3- How do instructional practice, teacher experience and education, classroom average achievement level and classroom SES composition mitigate students’ family background impact on their achievement?
Methodology, Methods, Research Instruments or Sources Used
The present study uses Swedish TIMSS 2019 data focusing on the science domains of biology, chemistry, and physics in the eighth grade. In Sweden, approximately, 4000 8th graders and 200 classes participated in TIMSS 2019, which are taught by an average of 3 science subject-teachers in each class.
The student questionnaire variable home educational resources was used as proxy of students’ socioeconomic status (SES). Teachers’ instructional practice, teacher’s years of teaching experience, completed level of formal education, and their major area of study were also selected from the teacher questionnaire data. The choice of the specific variables is justified by previous literature, indicating the influence of the included factors on student achievement.
The analysis is carried out simultaneously at student and classroom levels through two-level modelling that is used to investigate the effect of instructional practice, teacher experience, and teacher qualifications on differences in student achievement in biology, chemistry, and physics, which vary because of the provision of home educational resources. The application of multilevel analysis accounts for the potential cluster effects and allows for the investigation of the proposed research questions at the student and classroom levels. The sampling weight was used to make sure that the weighted sample matches the actual sample size in Sweden. The data management was carried out in IBM SPSS Statistics 29 and the models were estimated by Mplus 8.6 (Muthén & Muthén, 1998-2017). All five plausible values were used.
The two-level modelling technique was applied in a stepwise manner:
• Firstly, a Confirmatory Factor Analysis was carried out to test the validity of the construct instructional practice (IP).
• Next, a random slope-only model of the relationship between students’ family SES and their science achievement in each subject was run to test whether the relationship varies across different classrooms to decide upon the choice of the final model.
• If the random slope was not statistically significant, the latent variable IP was related to science achievement in a two-level random intercept model, controlling for student and class-level contextual characteristics (teachers’ experience and education, and classroom SES composition).
• If the random slope was significant, the compensatory effect of class-level factors on random slope was tested by regressing the random slope on the class-level factors in a two-level random intercept and random slope model. This was to account for the cross-level interaction between students’ family SES and their science achievement.
Conclusions, Expected Outcomes or Findings
The model fit indices suggested that the measurement model of Instructional practices fit the data well: X2(12) = 258.933, p = 0.00, RMSEA = .073 (90% CI = .065-.081), CFI = .946. SRMR = .033. Factor loading ranged from .45 to .71, indicating the measures of the latent constructs are valid and the measurement model can be established.
In the second step, random-slope-only models using the five plausible values for students’ biology, chemistry, and physics achievements and home educational resources were carried out. The results showed a significant variance of the random slope indicating that the relationship between student science achievements (biology and physics) and their home educational resources vary significantly across different classrooms.
Consequently, two-level random slope models using the data for biology and physics domains, and a two-level random intercept model using the data for chemistry domain were implemented. Interestingly, the results show that teachers' instructional practice has no significant influence on students’ achievement in biology, chemistry, and physics, when controlling for individual and classroom contextual characteristics. Teachers’ experience has a positively significant influence on biology achievement. However, it has no significant influence on chemistry and physics achievement at the eighth grade. In addition, teacher education and their major area of study had no significant influence on student achievement in the three domains of science. There are no significant relations between teachers’ experience, education, and classroom SES-composition and teachers’ instructional practice based on the Swedish TIMSS data. However, classroom SES-composition is a positively significant predictor of student achievement in all three domains. The results also show that teachers’ experience and their education are significantly correlated.
References
Act, S. E. (2010). Svensk författningssamling, Skollagen. [The Swedish Code of Statutes. Education Act] 2010: 800.
Blömeke, S., Olsen, R. V., & Suhl, U. (2016). Relation of student achievement to the quality of their teachers and instructional quality. Teacher quality, instructional quality and student outcomes, 2, 21-50.
Goe, L. (2007). The link between teacher quality and student outcomes: A research synthesis. National comprehensive center for teacher quality.
Hanushek, E. A., & Woessmann, L. (2012). Do better schools lead to more growth? Cognitive skills, economic outcomes, and causation. Journal of Economic Growth, 17 (4), 267–321.
Harris, D. N., & Sass, T. R. (2011). Teacher training, teacher quality and student achievement. Journal of public economics, 95(7-8), 798-812.
Lee, S. W., & Lee, E. A. (2020). Teacher qualification matters: The association between cumulative teacher qualification and students’ educational attainment. International Journal of Educational Development, 77, 102218.
Luschei, T.F., Jeong, D.W. (2018). Is teacher sorting a global phenomenon? Cross-national evidence on the nature and correlates of teacher quality opportunity gaps. Educational Researcher. 47 (9), 556–576.
Muthén, L. K., & Muthén, B. O. (1998). 1998-2017. MPlus user’s guide.
Scheerens, J., & Blömeke, S. (2016). Integrating teacher education effectiveness research into educational effectiveness models. Educational research review, 18, 70-87.
Kelly, D. L., Centurino, V. A. S., Martin, M. O., & Mullis, I. V. S. (2020). TIMSS 2019 encyclopedia: Education policy and curriculum in mathematics and science. Retrieved from Boston College, TIMSS & PIRLS International Study Center website: https://timssandpirls. bc. edu/timss2019/encyclopedia.Toraman, Ç. (2019). Effective teacher characteristics. Asian Journal of Instruction, 7(1), 1-14.
Varsakelis, N. C. (2006). Education, political institutions and innovative activity: A cross-country empirical investigation. Research Policy, 35(7), 1083–1090.
Paper
Teachers' Formal Qualifications and Instruction in Grade 4: Effects on Student Achievement in Grade 4 and 6
University of Gothenburg, Sweden
Presenting Author:Effective teaching is a multifaceted endeavour influenced by various factors. It extends beyond the mere possession of subject knowledge and teaching experience and is intricately tied to teaching methods (Darling-Hammond, 1997; Hudson et al., 2021; Leino et al., 2022; Shulman, 1987; Wharton-McDonald et al., 1998). International large-scale assessments (ILSA) such as the Trends in Mathematics and Science Study (TIMSS) and Progress in International Reading Literacy Study (PIRLS) have been widely used to establish links between teachers and their students in mathematics (e.g., Toropova et al., 2019), science (e.g. Fauth et al., 2019) and in reading (e.g., Johansson et al., 2015; Myrberg et al. 2019). However, research on teacher effects has yielded conflicting and inconclusive findings (Blömeke & Olsen, 2019; Coenen et al., 2018; Goe, 2007), partly due to the diverse methodological approaches employed in various studies. A major contributing factor is the lack of comparability and precision in defining teacher competence and teaching quality indicators. Furthermore, significant variations exist among countries in terms of the length, structure, and content of teacher education and instruction, necessitating country-specific analyses with accurate information on these specific features.
Investigating teacher effects on student achievement through large-scale data, such as PIRLS assessment data, presents distinct advantages. Firstly, large-scale assessments provide large samples, where whole classes of students are sampled, allowing comprehensive analysis of teacher effects across diverse student populations. Secondly, these assessments offer multiple measures for evaluating teachers and their teaching quality. However, a major challenge with ILSAs when estimating the impact of teachers on student outcomes is that we cannot account for students’ prior achievement. This complicates the task of isolating the direct influence of teachers on student learning outcomes.
In the present study, we address this limitation by utilizing the Swedish PIRLS 2016 sample to which additional register information from earlier and later grades was added. This means that we are not only able to account for prior achievement but also investigate long-term teacher effects on student performance. More specifically we are investigating the relationships between teachers’ reading specializations and the short-term and long-term impact of teachers’ reading comprehension practices in grade 4 on student performance in the PIRLS assessment and students’ subject grade in Swedish in sixth grade. We make use of scores from PIRLS, students’ national test results in grade 3 as well as subject grade in Swedish in grade 6.
Our research questions are:
- What are the relationships between teachers’ reading specialization and students’ PIRLS achievement and subject grade in Swedish in grade 6?
- What are the relationships between teachers’ reading comprehension activities and students’ PIRLS achievement and subject grade in Swedish in grade 6?
Methodology, Methods, Research Instruments or Sources Used
The present study utilizes data from the Swedish sample in PIRLS 2016, comprising 4525 students and 214 teachers. Beyond the standard PIRLS assessment information, the Swedish dataset offers noteworthy extensions: information on students’ subject grades and national test scores. This unique feature allows us to access both earlier and later performance data for students. As a result, the current design possesses two features not commonly found in traditional PIRLS design. First, the current design includes students’ prior achievement in third grade, using national test results in Swedish. Second, we can study the effects of teacher characteristics and instruction in both the short and long term. Given that the PIRLS assessment takes place in fourth grade, we can analyze teacher effects in the short term, as students have had their PIRLS teachers for approximately 7-8 months. Additionally, by utilizing achievement data from grade 6, we can address the long-term effects of reading instruction and teacher specialization, considering that students in Sweden typically have had their teacher for 2.5 years in grade 6.
As predictors we selected information on teachers’ specialization/s in reading pedagogy during teacher training, information about the time spent on language and reading instruction each week, as well as information about teachers’ classroom reading comprehension activities.
As student outcomes, we selected students’ reading achievement in the PIRLS 2016 and Swedish achievement in grade 6. PIRLS 2016 was conducted both on paper and online and we use the scores from the paper-based assessment. Achievement in grade 6 was collected from subject grades, a letter scale ranging from F-A which, however, was converted to a numerical scale ranging from 0-5. Because teacher effects on student achievement can result from initial differences in student achievement rather than teacher competence, we controlled for students’ prior achievement in grade 3. This measure stems from national tests scores which are ranging from 0-18 points.
The study employed a hierarchical design, treating students within classrooms as nested units. The study relied on multilevel regression to account for potential cluster effects that are due to the nature of the data (e.g., Hox, 2002). Sampling weights were used to account for the stratification. The main method was Structural Equation Modeling (SEM) with latent variables to investigate the relationships between the predictors and outcomes. We used Confirmatory Factor Analysis to model latent variables of specializations and reading comprehension activities. Data analysis employed SPSS 29 and Mplus version 8 software.
Conclusions, Expected Outcomes or Findings
Our findings indicate a positive and significant relationship between teachers’ specializations in reading pedagogy and students’ Swedish grades (when controlling for prior achievement, β = .29 (.09), p< .01). This suggests that teachers with a specialization in reading pedagogy significantly influence student achievement, irrespective of students’ initial achievement level. However, this relationship does not extend to the PIRLS assessment results in grade 4. This discrepancy may be attributed to the fact that most students in Sweden have had their new teacher for only 7-8 months when the PIRLS assessment is administered. As a result, it is reasonable to assume that the short-term effects of the teacher may not be evident for achievement in PIRLS.
Our initial investigations into the latent variable representing teachers’ reading comprehension activities did not reveal a significant relationship with the outcome variables. For this reason, we conducted further analyses to explore potential nonlinearities between reading comprehension activities and the two student outcomes, both with and without controlling for prior achievement. A significant curvilinear relationship was observed for teachers’ reading comprehension activities on PIRLS achievement and the Swedish grade. This implies that the relationship between reading comprehension activities and achievement was positive to a certain level, then declines. Further investigations of these relationships are needed.
Limitations
The teacher sample in PIRLS may not fully be representative of the entire teacher population in grade 4, however, the average years of teaching experience in our sample align with those of the total population. Another potential limitation could stem from ceiling effects within the measure of the prior achievement, as these may not adequately differentiate the highest performing students. However, the correlation to PIRLS achievement was relatively high.
References
Blömeke, S., & Olsen, R. V. (2019). Consistency of results regarding teacher effects across subjects, school levels, outcomes and countries. Teaching and Teacher Education, 77, 170-182. https://doi.org/10.1016/j.tate.2018.09.018
Coenen, J., Cornelisz, I., Groot, W., Maassen van den Brink, H., & Van Klaveren, C. (2018). Teacher characteristics and their effects on student test scores: a systematic review. Journal of economic surveys, 32(3), 848-877. https://doi.org/10.1111/joes.12210
Darling-Hammond, L. (1997). What matters most: 21st-century teaching. The Education digest, 63(3), 4.
Fauth, B., Decristan, J., Decker, A.-T., Büttner, G., Hardy, I., Klieme, E., & Kunter, M. (2019). The effects of teacher competence on student outcomes in elementary science education: The mediating role of teaching quality. Teaching and Teacher Education, 86, 102882. https://doi.org/10.1016/j.tate.2019.102882
Goe, L. (2007). The link between teacher quality and student outcomes: A research synthesis. National comprehensive center for teacher quality.
Hox, J. (2002). Multilevel Analysis: Techniques and Applications. Taylor and Francis. https://doi.org/10.4324/9781410604118
Hudson, A. K., Moore, K. A., Han, B., Wee Koh, P., Binks-Cantrell, E., & Malatesha Joshi, R. (2021). Elementary Teachers’ Knowledge of Foundational Literacy Skills: A Critical Piece of the Puzzle in the Science of Reading. Reading research quarterly, 56(1), S287-S315. https://doi.org/10.1002/rrq.408
Johansson, S., Myrberg, E., & Rosén, M. (2015). Formal Teacher Competence and its Effect on Pupil Reading Achievement. Scandinavian journal of educational research, 59(5), 564-582. https://doi.org/10.1080/00313831.2014.965787
Leino, K., Nissinen, K., & Sirén, M. (2022). Associations between teacher quality, instructional quality and student reading outcomes in Nordic PIRLS 2016 data. Large-scale Assessments in Education, 10(1), 25-30. https://doi.org/10.1186/s40536-022-00146-4
Myrberg, E., Johansson, S., & Rosén, M. (2019). The Relation between Teacher Specialization and Student Reading Achievement. Scandinavian journal of educational research, 63(5), 744-758. https://doi.org/10.1080/00313831.2018.1434826
Rutkowski, L., Gonzalez, E., Joncas, M., & von Davier, M. (2010). International Large-Scale Assessment Data: Issues in Secondary Analysis and Reporting. Educational researcher, 39(2), 142-151. https://doi.org/10.3102/0013189X10363170
Shulman, L. S. (1987). Knowledge and teaching: Foundations of the new reform. Harvard educational review, 57(1), 1-22. https://doi.org/10.17763/haer.57.1.j463w79r56455411
Toropova, A., Johansson, S., & Myrberg, E. (2019). The role of teacher characteristics for student achievement in mathematics and student perceptions of instructional quality. Education enquiry, 10(4), 275-299. https://doi.org/10.1080/20004508.2019.1591844
Wharton-McDonald, R., Pressley, M., & Hampston, J. M. (1998). Literacy Instruction in Nine First-Grade Classrooms: Teacher Characteristics and Student Achievement. The Elementary school journal, 99(2), 101-128. https://doi.org/10.1086/461918
Paper
Instructional Quality as Mediator and Moderator of the SES and Student Achievement Relationship. Insights from Swedish TIMSS 2019 Data
University of Gothenburg, Sweden
Presenting Author:The relationship between student socioeconomic status (SES) and achievement is apparent in almost every educational system across the world. In the Nordic educational systems, although it may be weaker than other countries, SES is yet one of the strongest predictors of academic achievement (Sirin, 2005). Few studies, such as Myrberg & Rosén (2009), explore the direct and indirect effects of various factors on the relationship between SES and student achievement. Further investigation into these mechanisms is necessary, as SES is mostly used to control for selection bias (Gustafsson, Nilsen, & Hansen, 2018).
Given the Nordic educational systems’ aim of increasing equity, an important apsect to investigate is to what degree teacher related factors influence the relationship between student SES and achievement. Previous studies have indicated that teachers account for a significant portion of variance in achievement between classrooms (Darling-Hammond, 2014). Indeed, there is a consensus that teachers are a crucial school factor, and their competence is the foundation of high-quality schools, instruction, and learning (Blömeke, Olsen & Suhl, 2016). Teacher competence (e.g., higher quality instruction, efficient classroom management etc.) positively affects student achievement (Kelcey et al., 2019) and overall may have a positive effect on reducing inequity in education (Wößmann, 2008).
Specifically, an essential characteristic for effective teachers lies in their ability to deliver quality instruction, by explaining the content clearly and assessing student understanding of the subject matter (Ferguson, 2012). While instruction quality is related with student motivation, it has been documented to be positively related with student achievement in mathematics (Bergem, Nilsen & Scherer, 2016). This is an important aspect, especially in Sweden, where challenges arise due to the unequal distribution of well-qualified teachers across schools, leading to a widening achievement gap between schools and student groups (Yang Hansen & Gustafsson, 2019). Further, not many studies have investigated the effect that instructional quality has on the relationship between student SES and achievement in Nordic educational systems. Also, a limited number of studies have utilized representative International Large-Scale Assessment (ILSA) data in this context.
Thus, the main objective of the present study is to investigate how instructional quality relate to equity in education. Specifically, the study focuses on how student perceptions of instructional quality may mediate or moderate the relationship between SES and eighth grade students’ math achievement in the Swedish educational system. This study is grounded in the Dynamic Model of Educational Effectiveness theory that explores factors influencing student outcomes across all school levels, which can be either equitably or inequitably distributed. The model acknowledges the nested structure of educational systems and the relationships among various factors at different levels. Specifically, the model refers to observable instructional behaviors of teachers in the classroom and includes eight instructional quality dimensions: orientation, structuring, questioning, teaching-modelling, application, time management, creating a learning environment, and classroom assessment (Creemers & Kyriakides, 2013). The research questions that guide the study are:
- To what extent does student perceived instructional quality and math achievement relate?
- Does classroom level student perceived instructional quality have an effect on the relationship between student SES and math achievement?
- To what extend does student level student perceived instructional quality mediates the relationship between student SES and math achievement?
Methodology, Methods, Research Instruments or Sources Used
The study uses cross-sectional secondary questionnaire data, to examine the relationship of SES and students perceived instructional quality with their math achievement. Particularly, it used the Swedish grade 8 data from the TIMSS 2019 cycle with a sample size of N=3996 Swedish students. Teacher instructional quality was measured using questionnaire indicators, such as ‘My teacher is good at explaining mathematics’, ‘My teacher has clear answers to my questions’, ‘My teacher links new lessons to what I already know’ etc. These items were measured on a 4-point Likert scale ranging from “agree a lot” to “disagree a lot”. For measuring student SES, information on the number of books at home and students’ responses regarding their father’s and their mother’s education was used. The number of books at home variable has been identified in several studies to be highly correlated with TIMSS achievement (e.g. Wiberg, 2019). A measure of student mathematics achievement, represented by five plausible values for each student’s math performance on a continuous scale provided by the IEA, was utilized in the analysis through a multiple imputation technique.
The method of confirmatory analysis was used to test whether the data fit the measurement models, and then there were built structural models based on an extensive literature review. Multilevel structural equation modelling techniques are employed in the study, as educational systems have an inherently multi-layered structure. Students are nested within classrooms, classrooms within schools, and the schools collectively form a national educational system. When individuals are clustered within natural occuring units (e.g., classrooms, schools, etc.), they share unique components that can affect their school performance. Therefore, multilevel models are essential for decomposing variance into its originating levels (Hox, 2002). The data analyses, which was conducted in SPSS 29 and Mplus 8, incorporated student weights, cluster, and the robust maximum likelihood estimator (MLR), while the model fit was assessed using both local and global fit indices.
Conclusions, Expected Outcomes or Findings
The preliminary analyses have resulted in well-fitting measurement models for students’ SES and student perceived teachers’ instructional quality latent constructs. The structural models examining the direct and indirect effects of student perceived teachers’ instructional quality on math achievement resulted in an overall good model fit. Model results confirmed that both SES and student perceived instructional quality at student and classroom level significantly relate with math achievement, consistent with prior research. Also, it was found that there is a significant indirect effect of students’ SES to their math achievement through teachers’ instructional quality at the individual level. Further, it was tested the interaction effect of teachers’ instructional quality in a multilevel model. A random slope on the relation between SES and math achievement was specified and teachers’ instructional quality at classroom level was found to have a significant interaction effect on this relationship.
While this study centers on teachers’ instruction quality and the connection between student SES and math achievement in Sweden, its results hold significance beyond the Swedish context. Concerns about educational equity and the importance of promoting effective teaching quality are prevalent across every democratic educational system. There is a global movement towards prioritizing equity in education (OECD, 2018), with a consistent emphasis on schooling as a key ‘equalizer’ among individuals of diverse backgrounds, crucial for countries adopting a preventative approach to economic inequality (Hanushek & Woessmann, 2015). Research on how teachers’ instruction quality influences the relationship of student socioeconomic background and academic performance sheds light on the on the pivotal role of teachers in addressing equity issues. Thus, further research is needed to examine how effective teaching contributes in fulfilling schools’ compensatory mission, mitigating the strong correlation between SES and achievement in Sweden and beyond.
References
Bergem, O. K., Nilsen, T., & Scherer, R. (2016). 7 Undervisningskvalitet i matematikk [7 Teaching quality in mathematics]. In Vi kan lykkes i realfag: Resultater og analyser fra TIMSS 2015, [We can succeed in science: Results and analyzes from TIMSS 2015] (pp. 120-136). Oslo: Universitetsforlaget.
Blömeke, S., Olsen, R. V. & Suhl, U. (2016). Relation of Student Achievement to the Quality of Their Teachers and Instructional Quality. In T. Nilsen & J. E. Gustafsson (Eds.), Teacher Quality, Instructional Quality and Student Outcomes (pp. 21-50). Springer.
Creemers, B., & Kyriakides, L. (2013). Using the Dynamic Model of Educational Effectiveness to Identify Stages of Effective Teaching: An Introduction to the Special Issue. The Journal of Classroom Interaction, 48(2), 4-10.
Darling-Hammond, L. (2014). Strengthening teacher preparation: the holy grail of teacher education. Peabody Journal of Education, 89, 547–561.
Ferguson, R.F. (2012). Can student surveys measure teaching quality? Phi Delta Kappa, 94(3), 24–28.
Gustafsson, J. E., Nilsen, T., & Hansen, K. Y. (2018). School characteristics moderating the relation between student socio-economic status and mathematics achievement in grade 8. Evidence from 50 countries in TIMSS 2011. Studies in Educational Evaluation, 57, 16-30.
Hanushek, E. A., & Woessmann, L. (2015). The knowledge capital of nations: education and the economics of growth. Cambridge: MIT Press.
Hox, J. (2002). Multilevel analysis : Techniques and applications. Mahwah, N.J.: Lawrence Erlbaum.
Kelcey, B., Hill, H. C., & Chin, M. J. (2019). Teacher mathematical knowledge, instructional quality, and student outcomes: a multilevel quantile mediation analysis. School Effectiveness and School Improvement, 30(4), 398-431.
Myrberg, E. & Rosén, M. (2009) Direct and indirect effects of parents´ education on reading achievement among third graders in Sweden. British Journal of Educational Psychology 79, no. 4, pp. 695-711.
OECD. (2018). Equity in education: breaking down barriers to social mobility. OECD publishing: Paris.
Sirin, S. R. (2005). Socioeconomic status and academic achievement: A meta-analytic review of research. Review of Educational Research, 75(3), 417–453.
Wiberg, M. (2019). The relationship between TIMSS mathematics achievements, grades and national test scores. Education Inquiry, 10(4), 328–343.
Woessmann, L. (2008). Efficiency and equity of European education and training policies. International Tax and Public Finance, 15, 199-230.
Yang Hansen, K., & Gustafsson, J.-E. (2019). Identifying the key source of deteriorating educational equity in Sweden between 1998 and 2014 International journal of educational research, 93, 79-90.