Can AI-Assisted Inquiry Enhance Students' Decision-Making Skills in Socio-Scientific Issues? A Three-Group Experimental Study on Climate Change

Synthesis: A three-group quasi-experimental pretest-posttest study examining whether AI-assisted inquiry strengthens students' decision-making on a climate-change socio-scientific issue (SSI). Climate change is a quintessential SSI — it rests on science but cannot be settled by science alone, forcing learners to weigh costs, values, and competing interests under uncertainty.

Design

The study compared three parallel groups so the contribution of the AI could be separated from the contribution of the inquiry itself:

1. An AI-assisted inquiry group, where a conversational AI partner supported students' structured reasoning
2. A non-AI inquiry group, which did the same inquiry without AI
3. A control group

Students' decision-making was scored against a rubric anchored in the structured reasoning that good scientific decisions require — naming drivers, reading data, weighing stakeholder interests, and planning actions.

Motivation and rationale

The authors drew on inquiry traditions rooted in action competence and education for sustainable development, which hold that the aim is to help learners deliberate and act, not merely to know. They designed the tasks, instruments, and rubric so that students recognized the climate problem as their own, directly addressing recurring obstacles in climate education: Misconceptions about AI, a sense that the problem is distant, and low feelings of Learner Agency.

What this means for practice

• Instructors. Run AI-assisted inquiry as a distinct condition from inquiry alone, not as a substitute for it: the AI group outgained inquiry-only peers on total decision-making (d = 0.69) and traditional instruction (d = 1.88).
• Instructors. Spend the AI-supported time on the steps students enter weakest, since gains were largest for monitoring and adaptive management (shift 1.07, d = 1.37) and generating alternatives (shift 0.98, d = 1.33).
• Curriculum designers. Keep the scenario ill-structured and value-laden — a climate response balancing science against cost, fairness, and obligations to future generations — so students must weigh stakeholder interests rather than find a defensible answer.
• Instructors. Score decision-making step by step with an explicit rubric: pooled baselines showed about 85% of responses at Emerging or Developing levels, with problem identification strongest (mean 1.99) and forward planning weakest (monitoring mean 1.57).
• Researchers. Report the class-level assignment and model clustering explicitly when evaluating AI-assisted inquiry at scale, since the reported between-group contrasts assume independent responses.

Limitations

• Intact classes, not individual students, were assigned to the three conditions (270 students, 90 per group, grades 8–11, mean age 14.6), so responses within a class or school may not be fully independent and clustering could inflate precision.
• Decision-making was measured from written worksheets scored against a rubric; worksheets capture reasoning imperfectly and may favor students who write fluently.
• The intervention was relatively short and the post-test followed soon after, so the durability and transfer of the gains remain open.
• One step did not separate the groups: post-test distributions for data collection and analysis did not differ reliably, indicating basic evidence-gathering improves fairly evenly regardless of condition.

Connected Concepts

• Science Education
• Inquiry-Based Learning
• Critical Thinking
• Generative AI
• Assessment
• Learner Agency
• Situated Learning
• Student Engagement
• AI in Education

Connected Articles

• AI-Supported Inquiry-Based Learning in Photosynthesis and Respiration: Implications for Sustainable Science Teacher Education — AI-Supported Inquiry in Photosynthesis and Respiration
• AI as a Partner in Learning about, Doing, and Engaging with Science: Vigilance as the Key to Productive Augmentation — AI as a Science Partner and Epistemic Vigilance
• Artificial Intelligence in Science and Chemistry Education: A Systematic Review — AI in Science and Chemistry Education (Systematic Review)
• A Scoping Review of AI for Sustainability and Sustainable AI in Higher Education — AI and Sustainability (Scoping Review)

Citation

Gousopoulos, D. (2026). Can AI-Assisted Inquiry Enhance Students' Decision-Making Skills in Socio-Scientific Issues? A Three-Group Experimental Study on Climate Change. International Journal of Advanced Multidisciplinary Research and Studies, 6(3), 1956-1966.