Truace
Subscribe

Junior high students in AI-supported course improved more than classmates in control group

Study title: PENGEMBANGAN COURSE TERINTEGRASI AI BERBASIS PROBLEM BASED LEARNING UNTUK MENINGKATKAN KEMAMPUAN PENERAPAN PROSEDUR DAN PEMECAHAN MASALAH SISWA SMP

Researchers built an Informatics course on the Moodle platform that mixed digital materials, problem-based activities, and AI support that gave explanations and summaries of learning material. It was aimed at junior high students and at linking concepts with procedures and real-world problems.

Read the study abstract

ABSTRACT Informatics education requires instructional designs that connect conceptual understanding with procedural application and contextual problem-solving. This study aimed to develop an Artificial Intelligence (AI)-integrated course based on Problem-Based Learning (PBL) to support these two competencies among junior high school students. The product was developed using the Moodle platform by combining digital learning materials, problem-based activities, and AI support in the form of explanations and learni…

By the Truvace desk1 min read1 sourceEvidence: Moderate · 64/100 — Controlled or validated comparison · n = 68 · 1 independent sourcePeer-reviewedTRV-2026-2341· cite & verify
PENGEMBANGAN COURSE TERINTEGRASI AI BERBASIS PROBLEM BASED LEARNING UNTUK MENINGKATKAN KEMAMPUAN PENERAPAN PROSEDUR DAN PEMECAHAN MASALAH SISWA SMP
Students of Baozhong Junior High School 20200114 by 褒忠國中 雲端網. CC BY 2.0 · https://creativecommons.org/licenses/by/2.0

In brief

The results suggest the design may help students learn computational thinking, but the evidence is only descriptive so far. The authors state that statistical effectiveness still has to be confirmed, so it is unclear whether the gap would hold up in formal testing or in other schools.

Main points

  1. The course combined Moodle-based materials, problem-based activities, and AI explanations and summaries.
  2. A needs analysis of 113 students found 23% for decomposition, 50% for pattern recognition, 35% for abstraction, and 44% for algorithms.
  3. The experimental and control groups each had 34 students and took pretests and posttests.
  4. Contextual problem-solving rose from 56.80 to 83.10 in the experimental group versus 57.20 to 71.40 in the control group.

The gain

Junior high students using the AI-integrated problem-based course raised procedural scores from 58.40 to 82.60 and problem-solving scores from 56.80 to 83.10, more than the control group's gains.

The rundown

The study aimed to support two competencies among junior high students: applying procedures in computational thinking and solving contextual problems. The product was developed on Moodle, with AI offering explanations and summaries of learning material.

Before development, the team ran a needs analysis with 113 students. It measured four computational thinking skills, with results of 23% for decomposition, 50% for pattern recognition, 35% for abstraction, and 44% for algorithms.

For evaluation, 34 students were in the experimental group and 34 in the control group. In procedural application, the experimental group improved by about 24 points and the control group by about 12 points, a descriptive comparison.

Sources

  1. Peer-reviewedEDUCATOR : Jurnal Inovasi Tenaga Pendidik dan Kependidikan2026-11-02

The debate