Transforming clicks into critical thinking: An AI-based media literacy program for children

Synthesis: Demir & Akar (2026) evaluate the pedagogical impact of an AI-enhanced critical media literacy program for fourth-grade children in a Turkish public primary school, using an explanatory sequential mixed-methods design with 36 students. An 18-hour intervention structured on the 5E instructional model integrated generative AI tools such as ChatGPT and Grammarly, producing large gains in students' critical media literacy alongside qualitative shifts toward digital safety, source verification, online Ethics, and critical media critique.

Key Findings

• Large quantitative gains in the AI-supported group: paired-samples comparisons for the experimental group showed significant improvement in media reading (+3.50), media writing (+1.67), and total media literacy (+5.17), all at p < .01, whereas the control group improved only modestly and did not significantly advance in media writing.
• Between-group comparisons confirmed the effect: post-test differences favored the experimental group with large effect sizes — Cohen's d = 1.12 (reading), 1.18 (writing), and 1.31 (total literacy) — positioning AI-supported instruction as a catalyst for analytical, productive, and reflective media skills.
• Qualitative themes revealed multidimensional development: analysis of interviews, student artifacts (posters, drawings, slogans), and classroom observations surfaced four themes — positive uses of media, awareness of negative aspects, contributions to critical media literacy, and student feedback on AI-supported instruction — showing cognitive, attitudinal, and behavioral transformation.
• Six domains of critical media literacy growth emerged: digital self-protection and data privacy, purposeful and responsible media use, safe communication and boundary awareness, critical evaluation and misinformation awareness, online risk awareness, and media ethics and digital citizenship.
• The 5E model operationalized AI as a pedagogical agent: AI tools were embedded phase-by-phase (e.g., ChatGPT for reflective questions and Q&A, Grammarly and Canva AI for content refinement, Padlet for peer feedback) rather than used as isolated add-ons, with activities aligned to the Turkish Language and Social Studies curricula for curricular coherence.

What this means for practice

• Instructors. Embed AI tools at every 5E phase rather than adding them on at the end: ChatGPT for reflective questions and Q&A, Grammarly and Canva AI for content refinement, Padlet for peer feedback. The 18-hour sequenced program, not the tools in isolation, produced the gains reported here.
• Instructors. Teach digital self-protection and data privacy as explicit content alongside critical evaluation: the growth themes that emerged ran from privacy and boundary awareness to misinformation and media ethics, not just bias spotting.
• Instructors. Anchor the activities in the existing curriculum so media literacy is not a bolt-on; this program was aligned to the Turkish Language and Social Studies curricula and fit inside regular fourth-grade instruction without restructuring.
• Administrators. Check device, connectivity, and support readiness before mandating AI-integrated media literacy, since the authors identify infrastructure and access disparities as a scalability barrier in under-resourced settings.

Limitations

• Thirty-six fourth-grade students (18 experimental, 18 control) in a single public primary school in Turkey's Aegean region, assigned to condition by purposive sampling based on availability and the school's infrastructural readiness rather than at random.
• The intervention lasted 18 hours, which the authors state may not capture long-term cognitive or behavioral change.
• Media literacy was measured with a 12-item, 3-point Likert self-report disposition scale (Never/Sometimes/Always), and the qualitative evidence came from student reflections, posters, and observations, so response bias is possible.
• The tools presuppose devices, connectivity, and digital-literacy support that the participating school already had; the authors flag that these disparities were not tested for scalability.

Citation

Demir, R., & Akar, C. (2026). Transforming clicks into critical thinking: An AI-based media literacy program for children. International Journal of Artificial Intelligence in Education, 36, Article 100002.