One of the challenges of modern education is the steadily increasing need to plan the learning process with consideration of students’ individual needs, which are becoming increasingly diverse in terms of cognitive, academic, and motivational characteristics. This need becomes particularly acute when students are required to complete tasks involving higher-order thinking skills. In final physics examinations, students particularly struggle with tasks requiring causal explanations, argumentation, and evidence-based justification. Similar difficulties are observed in most international examinations (within IB and AP programmes), as well as in the PISA assessment.

This study examines how the integration of the principles of the Complex Instruction method (Cohen & Lotan, 1997) and the principles of Differentiated Instruction (Tomlinson, 2014), combined in the form of Differentiated Complex Instruction (DCI), makes it possible to address different learning styles when working on tasks that require higher-order thinking skills. Within DCI, Complex Instruction supports equitable role distribution in heterogeneous classrooms, while differentiation provides access to tasks through varied formats and levels aligned with students’ readiness, without reducing academic rigor.

The DCI method is grounded in L. S. Vygotsky’s sociocultural theory, according to which the development of cognitive skills occurs through mediated social interactions. The hypothesis of this study is that, through equitable collaboration and individualized ways of engaging students in tasks, each learner gains the opportunity to make a meaningful contribution by engaging in deeper reasoning within the part of the task for which they are responsible.

This approach helps reduce gaps in physics learning by focusing on the development of argumentation and inquiry skills, which often arise when all students are given identical tasks that do not take individual differences into account. Although conducted in Kazakhstan, the study aligns with global educational priorities related to equity and innovation, particularly within the STEAM framework emphasised in European education. Thus, improving argumentation skills represents a universal priority for any educational system, regardless of country or culture, making the findings of this study relevant to the international educational community.

The main research question is formulated as follows: how does the Differentiated Complex Instruction method influence the development of students’ reasoning and argumentation skills in physics lessons? To gain a deeper understanding of this main research question, the following sub-questions were examined:

1. What changes occur in students’ causal and evidence-based reasoning?
2. How does students’ participation in group activities and their interaction with one another change?
3. How do teachers’ pedagogical practices change when applying DCI?
4. How sustainable are the results achieved by students with different levels of prior preparation?

The aim of this study is to determine the impact of Differentiated Complex Instruction on the development of reasoning and argumentation skills among 11th-grade students in the process of learning physics.

The study employed a quasi-experimental mixed-methods design over the course of a seven-week academic term at the Nazarbayev Intellectual School of Science and Mathematics in Shymkent. A total of 53 eleventh-grade students participated in the study (experimental group: n = 27; control group: n = 26), with the control group taught using instructional methods standard for the national education system. The intervention was implemented within the “Dynamics” unit, which included two major topics: “Elasticity” (Hooke’s law and Young’s modulus) and “Gravitational Field” (motion of spacecraft and energy conservation). The study used groupworthy tasks that included hands-on laboratory experiments (determination of Young’s modulus) and simulation-based materials (analysis of changes in the orbital altitude of the ISS), with task packages designed for different group roles. An assumption of the experiment was that each participant would perform their assigned role responsibly and that effective interaction within groups would take place.

To examine the effectiveness of the DCI method, a quasi-experimental design with mixed quantitative and qualitative methods was employed. A sequence of twelve 80-minute lessons using DCI was planned for the experimental group. The control group continued learning according to the instructional approach adopted in the Nazarbayev Intellectual Schools system, without systematic differentiation through role distribution in higher-order tasks. The study involved 53 eleventh-grade students (aged 16–17) from a single school that enrolls students selected through a competitive process from both urban and rural areas. The experimental group consisted of two classes taught by two teacher-researchers (16 and 11 students, respectively). The control group also included two classes of 13 students each, taught by two other teachers. The groups were initially heterogeneous, as they had been formed in advance. Ethical standards were respected at all stages of the study. DCI tasks were designed in a groupworthy-format, requiring interdependent interaction among students. The tasks included experimental comparisons of different methods for determining Young’s modulus, investigations of spring and rubber deformation, calculation-based tasks comparing the elastic properties of two different wires, identifying factors influencing gravitational acceleration on different planets, and analysis of changes in the orbital altitude of the ISS. In each case, students worked in groups of four and were able to choose roles aligned with their interests and strengths. These roles most often included «experimenter», «discussion moderator», «data analyst», and «presenter». Differentiation involved the provision of data in multiple formats (numerical data, graphs, and statements), as well as varying levels of scaffolding (core and supplementary data from which students could select what was necessary and disregard what was not). A key element of the Complex Instruction approach was the allocation of roles among students based on their learning styles, with teacher intervention to ensure a balanced level of contribution from each group member. Pre- and post-testing employed adapted versions of the Lawson test, with the original structure preserved (the first question requiring answer selection, the second requiring selection of justification) within a physics context (Cronbach’s alpha reliability coefficient α = 0.82). Data sources also included analysis of video recordings of students’ group work with coding of reasoning levels (inter-rater reliability κ = 0.78), teachers’ reflective journals, and teacher-completed checklists assessing the fidelity of Complex Instruction implementation. Quantitative data analysis included ANCOVA to determine changes in reasoning scores. Qualitative analysis was conducted through triangulation of data.

The Differentiated Complex Instruction method demonstrated high effectiveness in two main areas: improving reasoning and argumentation skills, and increasing the equity of active participation within group work. According to the results obtained, a statistically significant improvement in reasoning skills was observed (Cohen’s d = 0.72; p < 0.01). This improvement was most pronounced in the construction of causal relationships, compared with more moderate improvements in the control group. A Lawson test administered one month later showed the stability of improvements in causal reasoning skills, indicating the sustainability of the results. During group discussions, an increase in student equity of 35% was observed, reflected in reduced dominance by initially higher-achieving students and a more even distribution of contributions. The most notable gains were observed among students who initially demonstrated lower levels of engagement. These roles simultaneously served as scaffolds for their responses and participation in discussions. This indicates that the DCI method makes it possible to reduce achievement gaps without lowering academic expectations. Teachers reported more productive discussions during group work, as well as improvements in their own lesson-planning skills when using a differentiated approach. Pedagogical practice showed that during the study, and especially after its completion, teachers began to plan higher-order tasks in advance, organizing group work with explicit status management through role distribution. At the same time, teachers encountered difficulties in selecting high-quality resources and independently designing suitable tasks, pointing to the need for targeted methodological support. The limitations of this study include a relatively small sample size, the focus on a single subject area (physics), and the dual role of the teacher as both researcher and instructor, which may have influenced the accuracy of the results. Further research could examine the long-term effects of DCI and compare its impact across different subject areas.

Cohen, E. G., & Lotan, R. A. (1997). Working for equity in heterogeneous classrooms: Sociological theory in practice. Teachers College Press. Cohen, E. G., Lotan, R. A., Scarloss, B. A., & Arellano, A. R. (1999). Complex instruction: Equity in cooperative learning classrooms. Theory Into Practice, 38(2), 80–86. https://doi.org/10.1080/00405849909543836 Tomlinson, C. A. (2014). The differentiated classroom: Responding to the needs of all learners (2nd ed.). ASCD. Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes. Harvard University Press. Lawson, A. E. (2000). Classroom test of scientific reasoning: Revised paper. Journal of Research in Science Teaching, 37(5), 483–500. OECD. (2019). PISA 2018 results (Volume I): What students know and can do. OECD Publishing. https://doi.org/10.1787/5f07c754-en Mercer, N., & Littleton, K. (2007). Dialogue and the development of children’s thinking: A sociocultural approach. Routledge.