TIMSS Mathematics Research Articles

EXPLORING STUDENTS’ LEARNING STYLES AND MATHEMATICS PERFORMANCE AMONG LOWER SECONDARY SCHOOL STUDENTS

Based on the result in TIMSS Mathematics Achievement 2019 for student grade 8, it shows that mathematics performance has not yet achieved the desired position in the Malaysia Education Blueprint (2013-2025). Furthermore, the average mathematics score reported by PISA 2022 in Malaysia declined by 31 points compared to the achievement in PISA 2018. The impact from this result will affects the quality of education in Malaysia compared to other countries. Students' academic performance may be reflected to some extent based on their learning styles. As such, it becomes useful information to guide teacher’s instruction. Many studies have discussed on student learning styles; however, there are varying findings among different levels of learners. Therefore, this paper aims to investigate the learning styles and mathematics performance of lower secondary students (Form 1 and Form 2) in one of the schools in the Klang Valley by using cluster sampling. There were 147 respondents from Form 1 and Form 2 selected in this study. This study was conducted using a quantitative method; a set of VARK learning style questionnaire was distributed to the respondents, and mathematics performance results were taken from midterm test. The framework of the study adopted VARK learning style which has 30 items with seven dimensions that include visual, auditory, reading/writing, kinaesthetic, group and individual learning styles preference. The data analysed descriptive and inferential statistical techniques by using SPSS tool. The results revealed no significant difference in learning styles and mathematics performance between high and low mathematics achiever. However, upon examining each learning style individually, it was found that high mathematics achiever students exhibited significant difference in their learning styles. Furthermore, the study found there is no relationship between students' learning styles and mathematics performance across both high and low mathematics achiever groups.

TIMSS Matematik Alt Test Performansının Orta Ölçekli Bir Şehirdeki Yansıması

The aim of this study was to examine the performance of fifth grade students, who are studying in a middle-sized city in the Central Anatolia region, on an achievement test consisting of TIMSS mathematics subtest items, in terms of learning area, cognitive level, and gender variables. To achieve this aim, a quantitative research method was used, and a relational survey was conducted. Firstly, validity and reliability study were conducted with 154 fifth grade students for an achievement test created based on the distribution of learning area and cognitive level of the items announced by official institutions for the fourth grade level of TIMSS mathematics subtests. Then, the actual study was carried out with 333 fifth grade students. According to the results, the students showed the highest performance in items at the highest cognitive level, which is the reasoning level, in all three learning areas (numbers and operations, geometry and measurement, data). In terms of learning area, the students showed the highest performance in data, while the lowest performance was exhibited in numbers and operations. When the results were examined according to gender, it was found that there were significant gender differences in the scores belonging to the applying level of the numbers and operations learning area and the knowing and applying levels of the geometry and measurement learning area.

Predicting cross-national sex differences in large-scale assessments of students' reading literacy, mathematics, and science achievement: Evidence from PIRLS and TIMSS

Causes of sex differences in educational achievement have been controversially discussed in the extant literature. It has been speculated that differing prosperity and equality of opportunities may be linked to these differences, but conclusive empirical evidence for such effects is unavailable. Here, we present evidence for sex differences in international large-scale assessments of reading literacy, mathematics, and science across 16 cohorts from 1995 to 2019. Our analyses of PIRLS and TIMSS reading literacy, mathematics, and science achievement data (N = 3,999,062; 90 countries) showed consistent advantages for girls in reading literacy (d range: −0.02 to 0.66). For mathematics and science this pattern was less unambiguous, yielding non-trivial effects in both directions (d ranges: −0.44 to 0.36 and − 0.50 to 0.46, respectively). Sex differences in all three domains were more pronounced in more egalitarian countries (β range 0.16 to 0.20). Higher national prosperity and educational investment predicted larger sex differences favoring fourth grade boys in mathematics and science (β range: 0.07 to 0.39) and became less meaningful with increasing student ages (β range for eighth graders: 0.17 to 0.21). In all, our findings suggest that influences of economic macro-indicators on sex differences in educational achievement are differentiated according to subject, indicating larger sex differences in mathematics and science in more egalitarian and prosperous countries.

Why do girls do better? Unpacking South Africa’s gender gap in PIRLS and TIMSS

South Africa exhibits one of the largest pro-girl gaps in education, yet the reasons behind this gap are poorly understood. This paper analyses South Africa’s pro-girl gap in Grade 4 reading and Grade 5 mathematics achievement. I make use of Oaxaca-Blinder decomposition analysis to decompose the observed gender gaps into their explained and unexplained components, separately by school socio-economic quintile. This approach allows me to present new evidence that part of the country’s pro-girl gaps in reading and mathematics are due to girls progressing through the early grades faster than boys. I also find that South African girls score higher on the constructs aimed at tapping student attitudes toward reading (reading self-efficacy and enjoyment, and engagement in reading lessons), and that these differences contribute significantly to the pro-girl gap in PIRLS reading achievement. Interestingly, despite outperforming boys in the TIMSS mathematics assessment, South African girls do not display more positive attitudes toward mathematics. The results suggest much more needs to be done to understand the reasons behind boys’ disadvantage in the early grades, since much of the pro-girl achievement gap in Grade 4 reading and Grade 5 mathematics can be attributed the pro-girl advantage in grade completion in earlier grades.

Profiles of academically resilient students: An examination on TIMSS mathematics data

Profiles of academically resilient students: An examination on TIMSS mathematics data

A Comparative Study on Omani and Iranian Fourth-Grade Students’ Mathematics Achievement

The present study aimed to compare Omani and Iranian fourth-grade students’ performance in the TIMSS 2015 mathematics test. The TIMSS assessment sampling method was applied for selecting the statistical samples. Furthermore, the rule of gender and parents’ educational level were analyzed to compare boys' and girls’ performance in the TIMSS mathematics based on these factors. The results showed that there was a statistically significant difference between Omani girls’ and boys’ performance in the TIMSS 2015 mathematics assessment. Hence, the Iranian boys’ and girls' t-test results showed that no statistically significant difference was found between Iranian boys’ and girls’ mathematics performance. Moreover, the ANOVA analysis showed that there is a significant difference between parents’ educational level and students' mathematics achievement. Keywords: Mathematics, TIMSS, Gender, Parental Education DOI: 10.7176/JEP/13-11-01 Publication date: April 30 th 2022

Learning Large Q-Matrix by Restricted Boltzmann Machines.

Estimation of the large Q-matrix in cognitive diagnosis models (CDMs) with many items and latent attributes from observational data has been a huge challenge due to its high computational cost. Borrowing ideas from deep learning literature, we propose to learn the large Q-matrix by restricted Boltzmann machines (RBMs) to overcome the computational difficulties. In this paper, key relationships between RBMs and CDMs are identified. Consistent and robust learning of the Q-matrix in various CDMs is shown to be valid under certain conditions. Our simulation studies under different CDM settings show that RBMs not only outperform the existing methods in terms of learning speed, but also maintain good recovery accuracy of the Q-matrix. In the end, we illustrate the applicability and effectiveness of our method through a TIMSS mathematics data set.

Kuwaiti Students’ Achievements in Mathematics: Findings From the TIMSS Assessments: Reality and Reasons

Kuwait, has participated in international TIMSS test for many years, however, there has been little systematic effort to compile and assess changes between genders. This study attempts to look at the results of Kuwaiti students on the TIMSS mathematics assessments in general and according to gender in particular and the views of supervisors on the relatively low performance of Kuwaiti students. The study employed a mixed method approach in which data analysis of test results, IEA-issued documents, and statistics and official reports were used. Two focus group interviews were conducted with a convenience sample of nine educational supervisors from mathematics and science. The results show that the performance of Kuwaiti students in both the fourth and eighth grades was extremely low on the TIMSS mathematics assessments in general and in content areas in particular, since the first participation in 1995, even though the performance of eighth graders showed a slight improvement in 2015 in all areas. The TIMSS results also show that the higher the level of thinking that was assessed, the lower the performance of Kuwaiti students was. The data indicate that throughout all years, Kuwaiti girls outperformed boys, considering the slight improvement especially in 2015, yet both performances lagged behind international norms. Focus group transcript analysis reveals that supervisors perceived that students’ low performance on the TIMSS assessment test is related to a number of reasons as lack of interest in TIMSS test, unfamiliarity with TIMSS questions, and students’ weakness in the Arabic language. The paper concludes the need to systematically evaluate the TIMSS results, and develop interventions and a competent national curriculum in Kuwait.

Socioeconomic status and beyond: a multilevel analysis of TIMSS mathematics achievement given student and school context in Turkey

Socioeconomic status (SES) is considered a major predictor of student and school achievement. In most cases, SES is not malleable or available for manipulation to improve students’ learning and achievement. Therefore, we explored other student and school-related factors that may be malleable to reduce achievement differences between students and schools in Turkey. We used the TIMSS 2015 fourth-grade mathematics data and analyzed it using hierarchical linear modeling. We found that SES at both student and school levels is a dominant factor related to mathematics achievement and a much stronger predictor at the school level. Early literacy and numeracy activities, preschool education, intrinsic motivation, and engagement in instruction were found to be significantly and positively associated with higher achievement within schools. Similarly, among the school-level variables, we found significant and positive associations between schools’ mathematics scores and instruction quality and school readiness, in the presence of SES. We recommend that preschool education and early literacy and numeracy activities at home should be promoted. Additionally, although the findings of this study indicate possible SES-based school segregation; parents, teachers, and school leaders, as well as students, should work toward a more positive school climate to reduce achievement disparities due to SES.

Validity and reliability Measurement of TIMSS 2011 test in mathematics across school location and supervising authority in Jordan: الصدق والثبات في القياس لاختبار تمس (TIMSS 2011) في العلوم للصف الثامن عبر موقع المدرسة والجهة المشرفة في الأردن

The purpose of this study is to examine the measurement invariance of TIMSS mathematics test for eighth grade across school location and the supervising authority (Ministry of Education, UNRWA) in Jordan. To achieve this, the researcher used the analytical descriptive approach to analyze the test results for Jordanian 8th graders who participated in the TIMSS study in 2011. Specifically, the responses of 547 students to 34 items in booklet eight and nine were analyzed. Seven items were excluded because the skewness index is ​​greater than 2 and the kurtosis index is greater than 7. The data of the remaining 27 items were analyzed by using Confirmatory Factor Analysis (CFA) and Multiple Groups Confirmatory Factor Analysis (MGCFA) through the statistical program (AMOS 22). To judge the suitability of the test factorial structure the following goodness-of-fit indices in Structural equation modeling (SEM) were used: Chi Square (χ2) test was a statistically significant value of 468.167 and a good comparison index was CFI statistically significant value of 0.934 which is greater than 0.80 and the square of the mean root error approx. The results showed that there is no Configural invariance according to the location of the school (countryside, city), and the values ​​of the χ2 were a statistical function and its value is 822.858 and a good comparison index CFI is a statistical function and its value is 0.918 which is greater than 0.80 and the mean square root of the approximate error RMSEA is a statistic function and its value is 0.024 which is less than 0.08 and a root index The mean of the standard residual squares of SRMR is statistically significant and its value is 0.0408 which is less than 0.09. The results showed that there is no Metric (or weak) invariance according to the location of the school (countryside, city), the values ​​of the χ2 were a statistical function and its value is 861.321 and a good comparison index CFI is a statistical function and its value is 0.913 which is greater than 0.80 and the average square error of the approximate error RMSEA is a statistical function whose value is 0.024 and is less than 0.08 and an index The root mean standard SRMR squares are 0.0432 and less than 0.09. The results also showed that there was no Scalar (or strong) invariance according to the location of the school (countryside, city), and the values ​​of the χ2 were a statistical function and its value was 910.730 and a good comparison index CFI was a statistical function and its value is 0.903 which is greater than 0.80 and the average square of the approximate error root RMSEA statistically significant value and 0.025 which is less than 0.08 and an index The root mean standard SRMR squared is 0.0430 and is less than 0.09. The results showed that there is no Configural invariance in the supervising authority (Ministry of Education, UNRWA), and the values ​​of the χ2 were a statistically significant value of 833.660 and a good comparison index CFI was a statistical function with a value of 0.898 which is greater than 0.80 and an average square of the root error of approximate RMSEA statistically significant value of 0.028 which is less than 0.08 The mean index of the standard residual squares SRMR is statistically significant, and its value is 0.0351, which is less than 0.09. The results showed that there is no Metric (or weak) invariance for the supervising authority (Ministry of Education, UNRWA), and the values ​​of the χ2 were a statistically significant value of 890.903 and a good comparison index CFI was a statistically significant value of 0.883 which is greater than 0.80 and the square of the mean root error of approx. RMSEA statistically significant value of 0.029 which is less than 0.08 and the mean root index of the standard residual squares SRMR are statistically significant and have a value of 0.0620 which is less than 0.09. The results also showed that there is no Scalar (or strong) invariance of the supervising authority (Ministry of Education, UNRWA), and the values ​​of the χ2 were a statistical function and its value is 993.653 and a good comparison index CFI is a statistical function and its value is 0.854 and it is greater than 0.80 and the square of the average root error of RMSEA is a statistic function and its value is 0.033 which is less than 0.08 and the mean root index of the standard SRMR squares are statistically significant and their value is 0.0634 which is less than 0.09.

A Comparison of Diagnostic Profiles of Students in Korea and the USA on TIMSS Mathematics Assessment

A Comparison of Diagnostic Profiles of Students in Korea and the USA on TIMSS Mathematics Assessment

Образовательный выбор учащихся после 9-го и 11-го классов: сравнение первичных и вторичных эффектов социально-экономического положения семьи

В работе изучается связь характеристик семьи с выбором образовательной траектории в старших классах. Рассматриваются три ситуации образовательного выбора: выбор между высшим и профессиональным образованием после 9 или 11 класса и выбор вуза по окончании 11 класса. В соответствии с теорией Раймона Будона, эффекты социально-экономического положения семьи на образовательный выбор разделены на первичные и вторичные. Первичные эффекты проявляются в том, что учащиеся из семей с большим количеством различных ресурсов показывают более высокие образовательные достижения. Вторичные эффекты видны в том, что даже при одинаковой успеваемости учащиеся из семей с высоким социально-экономическим положением чаще продолжают обучение по сравнению со сверстниками из семей с низким социальным статусом. Работа основана на данных лонгитюдного исследования «Траектории в образовании и профессии», запущенного в 2011 г., когда респонденты учились в 8 классе, и продолжающегося до сих пор. Результаты показали, что первичные эффекты ослабевают при переходе из 9-го класса в 11-й. В то же время роль вторичных эффектов возрастает: даже при высокой успеваемости, учащиеся из семей с низким уровнем культурного, образовательного, социального капитала делают выбор в пользу простых образовательных стратегий. И наоборот, учащиеся из семей с высоким социально-экономическим положением, но невысокой успеваемостью делают выбор в пользу высшего образования, а не в пользу профессионального. На основе исследования предлагаются выводы о степени доступности образования в старших классах и университете, а также о вероятных причинах проявления неравенства. Предложены меры для преодоления социально-экономического неравенства на рассматриваемых этапах обучения.

Decoding the Meaning of Factorial Invariance and Updating the Practice of Multi-group Confirmatory Factor Analysis: A Demonstration With TIMSS Data

Measurement invariance (MI) has been developed in a very technical language and manner that is generally not widely accessible to social and behavioral researchers and applied measurement specialists. Primarily relying on the widely known concepts in regression and linear statistical modeling, this paper decoded the concept of MI in the context of factor analysis. The paper began by describing what is MI (and lack of MI) and how the concept can be realized in the context of factor analysis. Next, we explained the need for modeling the mean and covariance structure (MACS), instead of the traditionally applied covariance structure, in detecting factorial invariance. Along the way, we addressed the related matter of statistically testing for MI using the Chi-squared likelihood ratio test and fit indices in multi-group MACS confirmatory factor analysis. Bringing to bear current developments by Cheung and Rensvold (2002) and others, we provided an update on the practice of using change in fit statistics to test for MI. Throughout the paper we concretized our discussion, without lack of generality to other constructs and research settings, with an example of 21 crosscountry MI comparisons of the 1999 TIMSS mathematics scores.

An Application of Cognitive Diagnosis Modeling in TIMSS: A Comparison of Intuitive Definitions of Q-Matrices

Detection of students’ ability levels is one of the common aims in educational studies. Cognitive Diagnosis Modeling approach has been used recently for the purpose of ability level detection by defined Q-matrices. This paper aims to use Cognitive Diagnosis Modeling (CDM) in order to investigate the definition of a Q-matrix across the cognitive skills of different years and countries in TIMSS. There is a subjective way in defining Q-matrices; for this purpose, an application of building Q-matrices under specific CDMs, from a set of expert proposed attributes is examined. The proposed attributes are used to build Q-matrices for TIMSS mathematics questions across its cycles, and across different nations.

Comparing Results of TIMSS and the Hungarian National Assessment of Basic Competencies

In our paper, we compared some characteristics of TIMSS 2015 and the National Assessment of Basic Competencies (NABC) 2015. The NABC assesses all students’ reading and mathematics performance in Grades 6, 8 and 10. Both studies assessed Hungarian Grade 8 students’ mathematics abilities in the spring of 2015. We linked data of the two studies on the student level using Student Measurement IDs. We compared TIMSS and NABC mathematics scales based on the Assessment Framework of the two studies along with the results of students in the two assessments. The comparison of the Frameworks revealed that although the two tests use similar content and cognitive categorizations, there are crucial differences between the two constructs. While the basis of TIMSS’s mathematics construct is the common part of mathematics curricula of participating countries, NABC intends to measure mathematical literacy, the ability of students to use their mathematical knowledge and competencies in real life situations. The correlation between the TIMSS and NABC mathematics test results (0.79) also confirms that the two tests measure related, but not identical abilities. To evaluate the representativeness of the TIMSS sample we used school- and class-level weight factors of TIMSS and the student-level weights of NABC combined. The mean performances of the TIMSS sample are only slightly lower than the full NABC cohort’s, the effect size of the difference is 0.042 and 0.046 in mathematics and reading respectively. The differences in the standard deviations are somewhat but not considerably larger. The SES-index shows a very good match with no statistically significant differences in the mean and standard deviation of the sample and the full cohort. Our analysis confirms that estimations of population parameters based on TIMSS samples are of a good quality.

The TIMSS 2019 Item Equivalence Study: examining mode effects for computer-based assessment and implications for measuring trends

BackgroundTIMSS 2019 is the first assessment in the TIMSS transition to a computer-based assessment system, called eTIMSS. The TIMSS 2019 Item Equivalence Study was conducted in advance of the field test in 2017 to examine the potential for mode effects on the psychometric behavior of the TIMSS mathematics and science trend items induced by the change to computer-based administration.MethodsThe study employed a counterbalanced, within-subjects design to investigate the potential for eTIMSS mode effects. Sample sizes for analysis included 16,894 fourth grade students from 24 countries and 9,164 eighth grade students from 11 countries. Following a review of the differences of the trend items in paper and digital formats, item statistics were examined item by item and aggregated by subject for paperTIMSS and eTIMSS. Then, the TIMSS scaling methods were applied to produce achievement scale scores for each mode. These were used to estimate the expected magnitude of the mode effects on student achievement.ResultsThe results of the study provide support that the mathematics and science constructs assessed by the trend items were mostly unaffected in the transition to eTIMSS at both grades. However, there was an overall mode effect, where items were more difficult for students in digital formats compared to paper. The effect was larger in mathematics than science.ConclusionsBecause the trend items cannot be expected to be sufficiently equivalent across paperTIMSS and eTIMSS, it was concluded that modifications must be made to the usual item calibration model for TIMSS 2019 to measure trends. Each eTIMSS 2019 trend country will administer paper trend booklets to a nationally representative sample of students, in addition to the usual student sample, to provide a bridge between paperTIMSS and eTIMSS results.

Examination of TIMSS Mathematics Data with Multilevel Measurement Models in Respect to Content, Cognitive and Topic Areas

This research study aims to identify TIMMS 8th grade mathematics item groups and the specification of items in which Turkish 8th grade students have signıfıcantly lower level of correct responses compared to all other 8th grade participants. For this purpose, total 260 (82 from 1999, 88 from 2007, and 90 from 2011) items released by International Association for the Evaluation of Educational Achievement (IEA) were grouped according to cognitive, content and sub-content domains. Then, mean correct responses of released items for each participant country were obtained from IEA’s yearly almanac. Finally, data were analyzed by using Multilevel Measurement Models and differences in achievement levels between Turkish 8th graders and their peers from other participating countries were predicted and tested in the context of item groups. Analysis of data showed that performance of Turkish students statistically significantly lower than performance of students from rest of the other participant countries in Number (Content Domain)-Fractions and Decimals (Topic Area)-Knowing (Cognitive Domain) item group. Detailed investigation revealed that students generally fail in procedures in fractions and conversions among fraction, decimal, and percent.

A Comparative Study on TIMSS Mathematics Assessment of Korea, Japan, and the USA : A Differential Item Functioning Approach

A Comparative Study on TIMSS Mathematics Assessment of Korea, Japan, and the USA : A Differential Item Functioning Approach

An Analysis of TIMSS Mathematics Assessment Results Based on Multilevel Item Response Theory

An Analysis of TIMSS Mathematics Assessment Results Based on Multilevel Item Response Theory

Comparing DIF methods for data with dual dependency

The current study compared four differential item functioning (DIF) methods to examine their performances in terms of accounting for dual dependency (i.e., person and item clustering effects) simultaneously by a simulation study, which is not sufficiently studied under the current DIF literature. The four methods compared are logistic regression accounting neither person nor item clustering effect, hierarchical logistic regression accounting for person clustering effect, the testlet model accounting for the item clustering effect, and the multilevel testlet model accounting for both person and item clustering effects. The secondary goal of the current study was to evaluate the trade-off between simple models and complex models for the accuracy of DIF detection. An empirical example analyzing the 2011 TIMSS Mathematics data was also included to demonstrate the differential performances of the four DIF methods. A number of DIF analyses have been done on the TIMSS data, and rarely had these analyses accounted for the dual dependence of the data. Results indicated the complex models did not outperform simple models under certain conditions, especially when DIF parameters were considered in addition to significance tests. Results of the current study could provide supporting evidence for applied researchers in selecting the appropriate DIF methods under various conditions.