How Primary Teachers View ChatGPT: Developing an Attitude Scale for Math Education

The study develops and validates a reliable two-dimensional scale to measure primary school teachers’ attitudes towards using ChatGPT in mathematics education, focusing on its pedagogical potential and support for individualised learning. It shows that teachers’ beliefs, emotions, and intentions play a decisive role in how effectively AI tools like ChatGPT can be integrated into primary mathematics classrooms.

A study conducted within Türkiye’s public education system under the Ministry of National Education, with ethical approval from Necmettin Erbakan University and coordination by the Ankara Provincial Directorate of National Education, addresses a growing concern in global mathematics education: how teachers respond to artificial intelligence tools such as ChatGPT. As classrooms become more diverse and student achievement in mathematics declines worldwide, teachers are under pressure to personalise learning while managing heavy workloads. ChatGPT has emerged as a prominent tool promising instant feedback, interactive explanations, and support for individual learning. Yet its effectiveness depends not on technology alone, but on whether teachers trust it, feel confident using it, and see it as pedagogically meaningful.

Teachers at the Centre of AI Integration

The article argues that the successful integration of ChatGPT in mathematics education hinges on teacher attitudes. Drawing on well-established psychological theory, attitude is understood as a combination of beliefs, emotions, and behaviours. In practical terms, this means what teachers think about ChatGPT’s usefulness, how they feel about using it in class, and whether they intend to adopt it in their teaching. While previous studies have explored ChatGPT in education through interviews or small experiments, the author identifies a clear gap: there was no reliable quantitative tool to systematically measure primary school teachers’ attitudes towards using ChatGPT specifically in mathematics lessons. Filling this gap is the central purpose of the study.

Building a Reliable Attitude Scale

To develop this tool, the researcher followed a structured and transparent process. An initial pool of items was created by reviewing international research on mathematics education, artificial intelligence, and teacher attitudes. These items were reviewed by academic experts in AI and primary education to ensure clarity and relevance. The final draft consisted of 30 statements rated on a five-point scale, covering beliefs about usefulness, feelings such as excitement or anxiety, and classroom practices related to ChatGPT. A pilot study with 250 primary school teachers in Ankara tested the structure of the scale. Statistical analysis showed that the items worked very well together and could be grouped into two clear dimensions, explaining a large proportion of teachers’ responses. Importantly, none of the items had to be removed, indicating strong design from the outset.

What the Results Reveal

The main study, involving 300 teachers from different districts, confirmed the pilot findings. The analysis consistently showed two key dimensions shaping teachers’ attitudes. The first, described as perceived pedagogical potential, reflects how positively teachers view ChatGPT’s role in making mathematics lessons more engaging, reducing student anxiety, and supporting inclusive classrooms. This dimension captures both emotional reactions and behavioural intentions, such as willingness to receive training or enthusiasm about using new tools. The second dimension, support for individualised learning, focuses on teachers’ beliefs about ChatGPT’s ability to provide equal opportunities, assist learners at different levels, and help teachers manage diverse classrooms. Statistical tests showed that the scale is both valid and highly reliable, meaning it measures teacher attitudes consistently and accurately. While some indicators were only just within ideal thresholds, the overall evidence strongly supports the scale’s quality.

Implications for Schools and Policy

Beyond technical validation, the article places its findings within the wider debate on artificial intelligence in education. ChatGPT is portrayed neither as a cure-all nor as a threat, but as a powerful tool whose impact depends on thoughtful, teacher-led use. The scale developed in this study offers researchers, teacher educators, and policymakers a practical way to assess readiness for AI integration, identify training needs, and monitor changes in attitudes over time. The author recommends adapting the scale for use in different countries and cultural contexts, allowing comparisons across education systems. Although the study is limited to teachers in Ankara, the city’s demographic diversity strengthens its relevance. Overall, the article provides a clear, accessible tool for understanding how teachers engage with ChatGPT, reinforcing the idea that attitudes, not algorithms alone, will shape the future of mathematics education.