Abstract
Results of the TIMSS 2019 reveal that Kosovar students have lower achievement levels in both mathematics and science. In understanding this gap in achievement, it is of value to take a closer look at factors that may influence student learning. Results of this analysis reveal that factors such as home and school resources influence achievement with students who report more home resources and attend high-resource schools outperforming other groups of students. Student perceptions of their learning also shaped achievement with students performing higher when they had higher levels of confidence and preference for the subjects.
Introduction
The Trends in International Mathematics and Science Study-TIMSS is one of the most detailed and comprehensive education and achievement test in the world. The test assesses student achievement in math and science for a student attending fourth and eighth grade. In addition to test instruments, questioners are distributed to principals and teachers as well. The study provides national and cross-national data that can be analyzed by policy builders around the world. This assessment study is organized by a non-profit consortium of a research institute, the International Association for the Evaluation of Educational Achievement (IEA). The first TIMSS assessment was organized in 1995 and the last assessment in 2019 was the seventh assessment study, as studies are organized every 4 years and the upcoming assessment will take place in the spring of 2023 (Clerkin, 2018). According to IEA, a total of 64 countries participated in the 2019 assessment. In addition to assessing abilities in mathematics and science, TIMSS has a number of questionnaires that assess a wide variety of variables that are expected to impact student learning. One such variable is the socioeconomic status of the family.
Society, culture, political, and economic situations in the country play a role in student learning economic resources and equity in terms of division of socioeconomic resources links to favorable learning environments that lead to higher student achievement (Chiu & Khoo, 2005). Higher economic resources make it possible for schools to have more facilities and better-trained teachers, financial resources make it possible for countries to ensure equity by making materials and technology accessible bay all students (Bishop & Wössmann, 2004). Some countries tend to have centralized educational systems, which means that decisions are made at the national level. Such countries tend to have uniform policies, textbooks, and curricula. Many other countries on the other hand have decentralized systems, which means the policy decisions can be made by schools and local governments (Van de Werfhorst & Mijs, 2010). In decentralized educational systems, there are differences in how schools are organized and how they teach their students. Interestingly enough centralized systems tend to lead to higher rates of equality and higher students’ achievement (Bishop & Wössmann, 2004; Schneider et al., 2005; Van de Werfhorst & Mijs, 2010).
Socioeconomic status in TIMSS is measured by researching learning support materials that students have access at home. To that end students and parents are asked if the student has a room, study desk computer, tablet, and the mobile phone that can be used for learning (Broer et al., 2019). Additionally, students are also asked to identify if they have access to the internet and the number of books they have at home (TIMSS & PIRLS International Study Center, 2018). In TIMSS the home resources for learning scale, known as HRL, uses student and parent responses (Yin & Fishbein, 2020). On this scale students and parents alike are asked to know home resources and the answer varies from many, some, and few (Broer et al., 2019). Considering the importance of socioeconomic background on achievement, current research analyzes the link between socioeconomic indicators and student achievement in mathematics and science.
Literature Review
Arguably, knowledge of mathematics and science is vital to students’ future professional development, and the same importance links to the prosperity of society. TIMSS through its tests and context questioners gathers data and makes it available to everyone to enhance and support student learning in science and mathematics. Context questioners provide reach information about school structure, curricula teacher, and classroom practices, among others to contribute to the improvement of teaching and learning across countries (Koršňáková et al., 2022). This information can lead to more effective teaching and learning strategies across countries that participate in this international assessment program complete questionaries that required them to provide data on national policies as well as on national curricula for science and mathematics. Students, whose learning is assessed by TIMSS are enrolled in the fourth or eighth year of primary school, and the assumption is that what they learn in mathematics and science was learned in school or at home. To that ended school, classroom and home environments can be very effective learning venues. To assess these different factors, TIMSS includes questionnaires about the home contexts school context, classroom context, and finally attitudes characteristics of students (Koršňáková et al., 2022). Socioeconomic status can often be a form of inequality. Ferreira and Gignoux (2011), maintain that there are two forms of inequality. The first is inequality of outcomes, which refers to the difference in outcomes, and the second factor is inequality in opportunity which refers to the impact of family background and personal characteristics on student achievement. Studies have largely researched the link between SES and student achievement in an attempt to explain the role of socio-economic background (Lee & Burkam, 2002). Studies have measured the role of economics, social and cultural capital on the achievement and performance of students across generations (Bourdieu, 1986; Coleman, 1988, 1990; Lareau, 2011). It is untenably true that studies have documented the link between SES and academic achievement. What remains to be researched is how the education system influences this link by making it stronger or weaker. Understanding this association is important since most educations policies tend to address the achievement gap caused by SES and to improve achievement among disadvantaged students. TIMSS data that already spread across decide in enable researchers to explore and find answers about student achievement and predetermine factors such as SES. There is the consensus among researchers that SES is measured by parental education, income, and parental occupational prestige (Brese & Mirazchiyski, 2013). Gathering objective data in large-scale assessment is hard especially when asking students about family income and parent jobs. Students may simply not be able to provide accurate information this is the reason why largescale assessment asked students to report only the education level of parents. The argument here is that parental education influences all the other SES components since parents who have higher levels of education tend to have more prestigious jobs that carry higher incomes. Yang (2003) notes that home positions can be a better indication of SES level, such as the number of books available at home. According to Ogut and Broer (2016), the number of books at home is a good indicator of parental education level. In conclusion, while there are limits and differences in construct used to measure SES, it is crucial to assess the validity of the SES construct depending on the link to the student performance (Broer et al., 2017).
Home Resources for Learning
Home resources for learning are a measure of parental socioeconomic status along with the education level of parents. According to research, parental socioeconomic status such as education, income, occupation, and home resources are powerful predictors of students achievement (Bradley & Corwyn, 2002; Dahl & Lochner, 2005; Davis-Kean, 2005; Sirin, 2005). Nowadays, considering technology developments students are spending more time with digital media such as tablets and smartphones (Gutnick et al., 2011; Rideout et al., 2010). To that end, parents are reported to accept this behavior and approve of their children spending time in digital media because the underlying assumption is that by doing so students will become proficient in computers and technology and this is an important competence for academic and carrier development in the world of technology (Takeuchi, 2011). Literature, support this assumption to the degree that educational application can be effective and aid learning in young children (Chiong & Shuler, 2010; Lieberman et al., 2009).
School Composition by Student Socioeconomic Background
Literature suggests the collective socioeconomic status of all students that attend the school, influences achievement at the individual level (Rumberger & Palardy, 2005a; Sirin, 2005). Often, according to the research, schools that have many students with disadvantaged socioeconomic status can consider education and schooling as practices that have little influence (Agirdag et al., 2012). The stipulated link between lower achievement and low socioeconomic status can also be attributed to school factors, more specifically schools that have a higher number of low socioeconomic bedground students also tend to have a higher number of unqualified or low qualified teachers (Koršňáková et al., 2022). The correlation between lower socioeconomic status and lower achievement may be able to be partially explained by other school factors. For example, in some countries, schools with students from lower socioeconomic status are taught by less qualified teachers (Akiba et al., 2007a, 2007b). There is a universal understanding in research that the socioeconomic status of students influences achievement. There is ample evidence that student achievement in PISA is strongly influenced determined by the socioeconomic of the family i.e., parental education level and home positions. While the study documents the link between familial SES and the performance of 15 years’ student in reading, mathematics, and science no study assesses the link between these variables in 4th-grade students as measured by TIMSS. To that end, the main research question in this study is to what extent the socioeconomic background of the family influences the performance of 4th-grade students in TIMSS scales.
The hypotheses for the current study are presented below:
H1: There are gender differences in student achievement.
H2: More home resources for learning link to higher achievement of students in mathematics and science.
H3: Schools that have higher percentages of students from affluent socioeconomic backgrounds tend to have higher students’ achievement in mathematics and science.
Methodology
Current research makes use of the TIMSS Data Explorer available at the website of the IEA TIMSS & PIRLS International Study Center (IEA Explorer, 2019 available from https://www.iea.nl/data). The online data set provides information on all variables for all countries participating in the TIMSS assessment. In the case of this publication, only the data for Kosovo were analyzed. In 2019, 4754 students from 145 schools in Kosovo participated in the assessment. The data explorer allows researchers to conduct several statistical analysis and these analyses were done in the TIMSS data explorer which enables researchers to select variables and conduct tests on their relation to TIMSS results.
Table 1 below provides information on the overall performance of Kosovar students in the 2019 TIMSS assessment compared to other countries. Kosovar students scored 444 points in Mathematics and 413 points in Science, and according to this performance, Kosovar 4th grades scored 181 points lower in Math and 182 points in science when scores are compared to those of the highest performing country. Compared to the lowest performers, Kosovar students scored 147 points higher in mathematics and 164 points higher in science. In terms of gender differences girls (442 points) performed 5 points lower compared to boys (447 points), and in science girls (420) outperformed boys (407) by 13 points, a significant difference (p < .05).
Performance of Kosovar Students in the TIMSS Assessment Study.
In terms of performance distribution, only 1% of the participants from Kosovo were ranked in the advanced benchmark, another 8% were ranked in the high benchmark, 37% in the intermediate benchmark and finally the vast majority of students were identified as low performers in mathematics, 73%. Questions included in both scales, mathematics, and science, can measure knowledge in that particular domain, the competence to apply knowledge, and finally, the competence to use reasoning in mathematics and science. In the subscale of knowing and applying mathematics boys outperformed girls by 9 and 5 points, respectively. However, no difference between genders was found in reasoning in mathematics items. In science, gender differences are noticed across all subscales with girls outperforming boys by 13, 17, and 9 points, respectively (Table 2).
TIMSS data on home resources for learning reveals that when students and parents reported many resources for learning, students performed higher in both mathematics and science. The difference between students with many resources and students with some resources for learning was 55 points in mathematics and 68 points in science. The difference between students who had access to many resources and those who had access to only a few resources was even larger, it was 88 points in mathematics and 102 points in science. Among the Kosovar participants, only 4% reported having many resources for learning, the majority reported having some resources for learning 84% and 12% reported that they have very few resources for learning (Table 3).
Table 4 above presents information on student performance based on the socio-economic background of the student body. Among the participants 55% reported that they attend a more affluent school, 26% reported attending a neither more affluent nor more disadvantaged school, and finally, 19% reported attending a more disadvantaged school. Differences in performance can be noticed based on the attending school, with students performing the lowest in both science and mathematics when they attended a more disadvantaged school, followed by students attending a more affluent school. The highest performance is noticed in students who attend schools that are neither more affluent nor more disadvantaged. Table 5 below reveals that student absences are linked to performance. To that end, the low performers were the ones who reported the higher number of absences while the highest performers were the ones who reported that they are never or rarely absent. The difference between students who are absent once a week and those who are never or rarely absent is 53 points in mathematics and 48 points in science.
Student Distribution According to Benchmarks and Subscales of the TIMSS Study.
Home Resources for Learning Based on Student and Parent Responses.
School Composition by Socioeconomic Background of the Student Body.
Frequency of Student Absences.
Table 6 below presents student performance according to the degree to which students like the subjects. When students reported that they liked mathematics very much they outperformed students who reported somewhat liking mathematics by 55 points. The same holds for the performance in science where the highest performers were the students who liked science very much (429 points), followed by students who reported that they somewhat liked science (383 points) and the lowest performers were the students who reported that they did not like learning science topics (364). It is important to note that Kosovar students reported the highest preference for both topics, 78% of the participants reported that they liked math very much and 72% reported that they liked science very much. Higher percentages are noticeable only in the case of students from Albania in mathematics, and students from Albania and Portugal in science.
Performance in Mathematics and Science According to Student Preference for the Subject.
Confidence in learning the subject was revealed to be an influential factor in the performance of Kosovar students in TIMSS. Students who reported being very confident were the ones who had the highest scores in mathematics (473 points) compared to students who reported that they were somewhat confident (430 points) and students who reported not being confident in their abilities to learn mathematics (381 points). The difference in achievement is 43 and 92 points in mathematics. The same holds for achievement in science. Similar to mathematics, the students who reported being very confident in their ability to learn science had the highest performance (444 points), followed by students who reported that they were somewhat confident (404 points) and the lowest performers were the ones who were not confident in their ability to learn science (361 points). The difference between groups of students is 40 and 83 points, respectively.
Conclusion
According to Hoy (2002), they are three school characteristics that positively influence student achievement. These three factors are collective efficacy, collective reliability in parents and students, and finally the emphasis of the school on academics. A further factor that impacted achievement is school location, literature explains that schools differ depending on the student location, and students’ achievement is higher in urban schools compare to the schools located in rural areas. To that end, urban schools are bigger and have more resources, and do not have problems such as teaching shortages.
In addition, the urban school has higher percentages of qualified teachers, higher student teachers’ student-ratio compare to schools located in rural areas and smaller towns (OECD [Organization for Economic Cooperation and Development, 2006). The link between student achievement and family background is the universal trend. To that end, the strength of this relationship depends on the education system and social context of every single country. As such the achievement gap created by differences in backgrounds is more noticeable in the education system where inequality is high. as such education system should focus on providing quality by ensuring that the performance of students’ achievement is equal. This requires improving the scores of low SES students faster to address the achievement gap (Mullis et al., 2017).
In the case of Kosovo, students performed higher when they reported having more home resources compared to other groups of students. In Kosovo, the Ministry of Education Science and Technology (MEST) is the authority responsible for the pre-university education system as such within the ministry many divisions focus on assuring the quality of the pre-university of education such as the education inspectorate or the Division for Quality Assurance, Standards, Assessment, and Licensing (DQASAL). This division within the Ministry is responsible for developing policies as well as monitoring and conducting national and international assessments (Broer et al., 2019). The Ministry is also responsible for the distribution of resources in schools and other financial implications. The results of this analysis are important as based on TIMSS data it reveals that there are differences in terms of school resources and schools in the country can be categorized into three categories. To that end, students performed significantly higher when they attended schools that have a lot of resources compared to cases when they attended schools with fewer resources.
Finally, the results of the analysis also reveal that student perceptions may shape achievement and greatly contribute to the achievement gaps. Evidently, students who showed a higher preference for the subjects were the students with the highest performance both in mathematics as well as science. Additionally, confidence levels also revealed to be important since higher confidence in abilities to learn mathematics and science related to higher achievement in both scales. In conclusion, it is important for future research to understand to what degree teacher practices and school policies may impact student performance. It is of paramount importance to understand which students feel that they `belong` in the Kosovar education system in order to reveal hindering practices which contribute to lower student achievement and learning (Table 7).
Performance in Mathematics and Science According to Student Confidence in Learning the Subject.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
