Educational Gap Among Fourth Graders: Subject-Specific Strength and Weaknesses by Country, Gender, and Language Background

• In Event: Unequal Opportunities and Unequal Outcomes: International Evidence

Abstract

Objective
The paper aims at identifying subject-specific strengths and weaknesses in student achievement and how these performance patterns vary by gender, country, and language background by analyzing performance in TIMSS and PIRLS 2011. Although previous literature reported on the individual differences of learners, few researchers have focused on comparison of student learning outcomes across different subjects and their background characteristics. However, cross-subject comparison provides a fundamentally different perspective on student performance and thus has the potential to bring new insights into the development of curriculum and instructional design in various countries.
Theoretical Framework
According to the ecological system theory, individuals are embedded in the surrounding environment that influences their development (Bronfenbrenner, 1979). Thus, the macrolevel of the environment can refer to the curricula that emphasize particular subjects leading to the development of academic strengths or weaknesses among students. Another example is the societal views on gender roles. As noted by Hubbard (2005), some community cultures construct gendered attitudes among students and affect their educational aspirations and learning strategies. Moreover, family characteristics such as the language of communication at home could be extremely important for learning progress, especially in reading (Gersten & Woodward, 1994; Moss & Puma, 1995).
Data and Methods
Factor mixture analysis (FMA) was applied to combine latent continuous and latent categorical variable modelling to investigate the dimensionality of the student test scores in mathematics, reading, and science using the data from the 35 countries and regions that participated in TIMSS and PIRLS 2011. The estimated model parameters of the FMA model were used as fixed parameters, and each student was assigned to their most likely latent profile. We subsequently compared the distribution of the emerged performance profiles across countries, gender, and language background.
Results
Our results show that boys are slightly overrepresented in the profile with strength in mathematics and weakness in reading, while girls are overrepresented in the profile with a balanced performance across domains.
The study provides compelling evidence for cross-country differences in student performance. In some countries, almost all students were assigned to the profile with the balanced performance across domains (e.g., Austria, Spain, Oman), but in other countries (e.g., Iran, Italy, Ireland), a large share of students showed a particular strength in mathematics coupled with a pronounced weakness in science or reading.
The share of students in the profile with relatively low performance in reading increases with the decrease of the use of test language at home. In contrast, students who never speak the test language at home did particularly well in mathematics.
Conclusion
This study suggests that educational gap among students is more evident in their test results in mathematics and reading rather than in science, and their performance is interrelated with their background characteristics. These findings point to the importance of national curricula in shaping student performance in different domains, the need to design instructional strategies to address the gender gap in academic performance as well as support those students who require additional assistance in language acquisition due to their family background.

Authors

• Olesya Gladushyna, TU Dortmund University

• Rolf Strietholt, TU Dortmund University

• Isa Steinmann, TU Dortmund University