Topic: AI chatbots in STEM education
The study is designed to address the following research questions:
RQ1: What is the relationship between students’ use of AI bots and academic achievements in STEM subjects (Biology, Chemistry, Physics, Math) when controlled for such variables as gender, technology literacy, extracurricular activities, level of teachers’ support, self-directed learning, and access to resources?
RQ2: What is the relationship between students’ use of AI-bots and academic motivation in STEM subjects (Biology, Chemistry, Physics, Math) when controlled for such variables as gender, technology literacy, extracurricular activities, level of teachers’ support, self-directed learning, and access to resources?
RQ3: How do students use chat-bots to facilitate their independent learning in STEM subjects (Biology, Chemistry, Physics, Math)?
Objective: This study investigates the impact of AI-bot usage on students’ academic achievements and motivation in secondary STEM education, with a specific focus on Kazakhstan as an emerging AI-in-education context.

Theoretical framework:
The study is grounded in Self-Determination Theory (STD) (Deci & Ryan, 1985) and Social Cognitive Theory (SCT) (Bandura, 1986; 1999). SDT explains how AI-supported learning environments may enhance intrinsic motivation by supporting autonomy, competence, and relatedness (Deci & Ryan, 2000), while SCT highlights the role of self-efficacy in shaping academic engagement and performance (Schunk & DiBenedetto, 2020). Together, these frameworks allow for an integrated analysis of cognitive, motivational, and contextual factors influencing learning outcomes in AI-mediated environments.
While most empirical research on AI chatbots in education is concentrated in Western or East-Asian contexts, evidence from Central Asia remains limited. This study contributes internationally relevant data from Kazakhstan, offering comparative insights for European education systems facing similar challenges related to ethical AI use, equity, academic integrity, and policy alignment.

This study employed a quantitative, non-experimental correlational research design to examine the relationship between students’ use of artificial intelligence (AI) chatbots, academic achievement, and academic motivation in STEM subjects. This design was selected to explore natural associations between variables without manipulation, allowing for the analysis of relationships among AI usage, educational outcomes, and motivation while accounting for multiple confounding factors. The research was conducted at a private K–11 school in Kazakhstan. The site was chosen due to the researcher’s access and the school’s openness to educational innovation, including the integration of AI tools. Private schools in Kazakhstan often adopt new technologies more readily and provide individualized learning environments, making this context appropriate for investigating AI-supported learning practices. A convenience sampling strategy was used to recruit participants who were readily accessible and willing to participate. Although this non-probability sampling method limits the generalizability of findings, it allowed for in-depth examination within a real educational setting. The study acknowledges these limitations and recommends cautious interpretation beyond similar contexts. Data were collected using a structured survey instrument. The primary tool was the Science Motivation Questionnaire II (Glynn et al., 2011) which aligns with Self-Determination Theory and Social Cognitive Theory by measuring intrinsic motivation, extrinsic motivation, self-determination, self-efficacy, career motivation, and grade motivation. Additional survey sections gathered demographic information, self-reported academic achievement in STEM subjects, frequency and purposes of AI chatbot use, and contextual variables such as technology proficiency, extracurricular participation, teacher support, self-directed learning, and access to learning resources. Data collection followed strict ethical procedures and was approved by the Nazarbayev University Institutional Review Ethics Committee (IREC). Written parental consent and student assent were obtained, participation was voluntary, and anonymity and confidentiality were maintained throughout the study. Data analysis was conducted using Jamovi (version 2.3.21). Descriptive statistics were used to summarize participant characteristics, AI usage patterns, academic performance, and motivation levels. Multiple linear regression analyses were employed to examine the predictive relationship between AI chatbot usage and both academic achievement and motivation while controlling for confounding variables. Overall, this methodological approach enabled a systematic and ethically sound investigation into the role of AI chatbots in supporting STEM learning, providing empirical evidence relevant to both national and international discussions on AI integration in education.

The results of the study indicate that AI chatbot usage was not a significant predictor of either academic achievement or academic motivation in STEM subjects. For academic performance (RQ1), gender, technology literacy, and participation in extracurricular activities emerged as significant predictors, while AI use itself showed no statistically significant effect. Similarly, the analysis of academic motivation (RQ2) revealed no direct relationship between AI chatbot usage and student motivation. Instead, extracurricular activities demonstrated a strong and significant positive impact on students’ motivation. Regarding students’ patterns of AI chatbot use for independent learning (RQ3), descriptive findings showed that learners actively employed AI tools for a wide range of academic purposes, from routine tasks such as homework support to more complex activities including essay writing and concept simplification. Usage patterns varied by gender and grade level: male students reported higher use for conceptual explanations and learning beyond the classroom, while female students demonstrated higher overall engagement. AI chatbot use increased progressively from Grade 7 to Grade 10 and slightly declined in Grade 11. These results suggest that AI chatbots function most effectively as supportive learning tools rather than primary drivers of performance, emphasizing the importance of pedagogical integration and contextual factors. The study provides evidence-based guidance for educators and policymakers on responsible and effective AI implementation in STEM education.

Bandura, A. (1986). Social foundations of thought and action. Englewood Cliffs, NJ, 1986(23-28), 2. http://surl.li/sbizk Bandura, A., Freeman, W. H., & Lightsey, R. (1999). Self-efficacy: The exercise of control. https://doi.org/10.1891/0889-8391.13.2.158 Deci, E. L., & Ryan, R. M. (1985). Intrinsic Motivation and Self-determination in Human Behavior. NY: Plenum Press. https://doi.org/10.1007/978-1-4899-2271-7 Deci, E. L., & Ryan, R. M. (2000). The" what" and" why" of goal pursuits: Human needs and the self-determination of behavior. Psychological inquiry, 11(4), 227-268. https://doi.org/10.1207/S15327965PLI1104_01 Glynn, S. M., Brickman, P., Armstrong, N., & Taasoobshirazi, G. (2011). Science motivation questionnaire II: Validation with science majors and nonscience majors. Journal of research in science teaching, 48(10), 1159-1176. https://doi.org/10.1002/tea.20442 Schunk, D. H., & DiBenedetto, M. K. (2020). Motivation and social cognitive theory. Contemporary educational psychology, 60, 101832. https://doi.org/10.1016/j.cedpsych.2019.101832