Metin Ünal, Halime İrem Türkmen
9 min
As artificial intelligence (AI) continues to reshape educational landscapes, understanding the competencies of future educators is critical. This study investigates the current levels of AI literacy and AI readiness among 164 pre-service mathematics teachers. By examining the relationship between these two constructs, the authors aim to identify how well-prepared future mathematics instructors are to integrate AI into their pedagogy and what factors influence that readiness.
Using a cross-sectional survey design, the researchers collected data from pre-service teachers through standardized instruments: the AI Literacy Scale and the AI Readiness Scale. These scales measure specific dimensions including cognitive understanding, technical usage, ethical judgment, and vision for AI in education. Data were analyzed using descriptive statistics, t-tests, and ANOVA to identify differences across demographic variables (such as gender and year of study) and usage patterns. Finally, linear regression was employed to determine the predictive capacity of AI literacy for AI readiness.
The results indicate that pre-service mathematics teachers possess high levels of both AI literacy and AI readiness. While demographic variables like gender and grade level did not significantly influence these scores, behavioral patterns were critical. Participants who frequently use AI in their daily lives and those who actively follow AI developments reported significantly higher levels of both literacy and readiness. Furthermore, a strong positive correlation exists between AI literacy and AI readiness, with regression analysis confirming that AI literacy is a robust predictor of readiness, explaining approximately 55% of the observed variance. These findings underscore the necessity of formal training, as exposure and engagement appear to be primary drivers of professional preparedness.
For mathematics educators, AI offers transformative potential for personalized learning, automated assessment, and interactive problem-solving. This study suggests that if teacher preparation programs want to ensure their graduates are capable of leveraging AI effectively, they must move beyond general awareness and provide structured, curriculum-embedded training that directly enhances AI literacy.
The purpose of this study is to examine the level of artificial intelligence readiness and artificial intelligence literacy of pre-service mathematics teachers in relation to various variables, and to identify the relationship between these skills. The present study's sample comprised 164 pre-service mathematics teachers. The "Artificial Intelligence Readiness" and "Artificial Intelligence Literacy" scales were utilised as data collection tools. Statistical Package for the Social Sciences (SPSS) software was then deployed in order to facilitate data analysis, with the following analyses being conducted: independent sample t-tests, one-way analysis of variance (ANOVA), correlation analysis and regression analysis. The findings of the study demonstrated that pre-service mathematics teacher candidates exhibited a high level of both artificial intelligence readiness and literacy. Moreover a high level of positive and significant statistical relationship is demonstrated between artificial intelligence literacy and readiness. It was revealed that artificial intelligence literacy is responsible for explaining approximately 55% of the variance in artificial intelligence readiness.
Sam: Let's make the method concrete. What did a participant actually do, beyond saying whether they liked AI?
Alex: Participants completed questionnaires during classes, with researchers present to answer questions. They reported their habits and attitudes, then rated statements about literacy and readiness. Both scales had previously been developed and adapted into Turkish.
Sam: What kind of statement captures literacy without just measuring enthusiasm?
Alex: One asks whether someone can choose an appropriate solution from alternatives provided by a smart agent. That's about evaluating the output, not merely using the tool. Other statements cover recognising AI in products, using applications skilfully, and remaining alert to misuse.
Sam: And readiness shifts that judgment into teaching? I'd want to know whether it asks about classroom practice rather than general confidence.
Alex: It does ask about teaching. For example, one statement concerns using AI rationally to solve problems discovered during the teaching process. The readiness scale also covers understanding teachers' roles, thinking about educational opportunities and limitations, and recognising ethical responsibilities.
Sam: But those are statements about what respondents believe they can do. They weren't asked to select an actual solution or teach an actual lesson?
Alex: The reported data come from their ratings, not observed performance on those tasks. That's central to interpreting the findings. High scores mean strong reported literacy and readiness according to these instruments, not demonstrated success integrating AI into mathematics lessons.
Sam: With that boundary in place, what is the main result we should hold onto?
Alex: Both average scores fell into the authors' high category. More consequentially, the correlation between literacy and readiness was about point seven four. In this sample, candidates reporting stronger literacy generally also reported stronger readiness.
Sam: The headline also says literacy explains readiness. What does that mean here, and what would be reading too much into it?
Alex: Their regression model accounted for about fifty-five percent of the variation in readiness scores using literacy scores. That's a substantial statistical relationship within these responses. It does not mean literacy caused that share of readiness, or that a literacy course would produce an equivalent improvement.
Sam: So the correlation and the regression are two descriptions of the same underlying link, not independent confirmations?
Alex: Yes. The model uses literacy as its sole predictor of readiness. It puts a predictive description on the association, but it adds no intervention or sequence over time that could establish direction.
Sam: Were there comparisons that help distinguish literacy from readiness? Otherwise, the strong link might be the whole story.
Alex: There were useful differences. Candidates who frequently used AI in educational settings reported higher readiness than occasional or rare users. But literacy did not differ significantly across those educational-use groups.
Sam: That suggests classroom-related use and broader literacy aren't interchangeable. Did the same split appear in attitudes or interest in training?
Alex: Candidates willing to take an AI course had significantly higher readiness, but not significantly higher literacy. A positive attitude towards AI showed the same pattern. Those findings separate willingness and enthusiasm from the broader literacy measure, even though the two overall scores were strongly associated.
Sam: What did go with higher scores on both measures?
Alex: Daily-life AI use did. Frequent and occasional users scored higher on both measures than rare users. Regularly following AI developments was also associated with higher literacy and readiness.
Sam: It's tempting to turn that into advice to use AI more and follow the news. Does the design justify that prescription?
Alex: Not as a tested route to improvement. More literate, more prepared candidates might choose to engage more, or engagement might support those qualities. The survey cannot distinguish those directions. It also found no statistically significant differences by gender or year of study.
Sam: Let's put the strongest limitation plainly. How far should someone outside this university carry these results?
Alex: The sample was selected for convenience and accessibility at a single institution. It consisted only of mathematics teacher candidates. I'd treat this as evidence about that group, not an estimate for all Turkish teachers, other specialisms, or teachers already in classrooms.
Sam: And even within that group, similar questionnaire content could contribute to the strong relationship, couldn't it?
Alex: Both instruments include ethics and statements about ability to use AI. That gives a reason to consider conceptual overlap when interpreting the correlation. The excerpt does not establish how much overlap contributed, so we shouldn't assign it an effect.
Sam: What do the authors recommend teacher educators do, given those boundaries?
Alex: They recommend dedicated AI courses or lessons during undergraduate preparation. About seventy-seven percent of participants expressed willingness to take such courses. That's evidence of interest in this sample, while the recommendation's effectiveness remains untested here.
Sam: Who should read the full document, and where would you send them first?
Alex: Teacher educators planning AI instruction, and researchers measuring teacher competencies, should read the Data Collection Tools section first. It shows what literacy and readiness actually mean in this study. Then examine the contrasting results for educational use, course willingness, and attitudes, before reading the regression findings.
Sam: And what's the line to remember for someone who doesn't need the full measurement detail?
Alex: Among these future mathematics teachers, reported AI literacy tracked reported readiness strongly. Knowing, judging, and feeling prepared belong together—but this survey cannot tell us which builds which.
Sam: Keep that distinction in mind when planning teacher preparation.