Machine Learning Competence in Islamic Education Teachers Directly Boosts Student Academic Achievement, Study Finds

Gloria Florensia Br Surbakti Senin, Juli 20, 2026
Ilustrasi by AI

FORMOSA NEWS


A groundbreaking study conducted by researchers from Sunan Gresik University demonstrates that teachers' proficiency in machine learning technology directly drives significant improvements in student academic achievement within Islamic Religious Education (PAI). The investigation, carried out between February and June 2026 at SMP IT Al Uswah Tuban in East Java, evaluates how the integration of modern digital platforms shapes modern educational quality. As digital learning tools become commonplace in global academic environments, these findings prove that an educator's capacity to manage and apply predictive data analytics yields measurable benefits for student learning outcomes.

The Background of Digital Pedagogy

Modern academic ecosystems are experiencing a major paradigm shift driven by the rapid evolution of Artificial Intelligence (AI) and machine learning tools. In contemporary settings, educational institutions increasingly rely on adaptive, data-driven frameworks to monitor student metrics, personalize training materials, and identify pupils at risk of academic failure.

While digital tools are frequently used to handle basic institutional administration, their integration as core pedagogical instruments remains limited in many schools. This research bridges a critical literature gap by examining how structured machine learning skills enable teachers to transition from conventional content deliverers into strategic designers of intelligent, personalized educational experiences.

Research Methodology

The researchers from Sunan Gresik University employed a sequential explanatory mixed-methods approach to investigate these dynamics. The data collection involved a total sampling population consisting of 8 active Islamic Religious Education teachers and 124 students across grades VII through IX at SMP IT Al Uswah.

  • Quantitative Phase: Teachers completed a 40-item Likert scale questionnaire adapted from the global DigCompEdu framework, measuring skills across four core areas: data literacy, learning analytics, adaptive content personalization, and predictive pedagogical interventions. The researchers evaluated student academic success using official report card marks from the second semester of the 2025/2026 academic year alongside formative quiz tracking scores. Quantitative analytics were processed using structural equation modeling via SmartPLS 4.0 software.
  • Qualitative Phase: To provide empirical context, the authors performed detailed thematic analysis using data gathered from 12 separate classroom observations, in-depth interviews with 6 primary informants, and academic documentation reviews.

Key Findings

The structural equation modeling revealed a highly significant, positive relationship between teacher machine learning competence and student learning outcomes. The primary data points include:

  • Strong Direct Correlation: The statistical analysis demonstrated a high path coefficient ($\beta = 0.612, t = 7.84, p < 0.001$), proving that higher teacher competence leads directly to better student scores.
  • Variance Explanation: The structural model achieved an $R^2$ value of 0.462, meaning that teacher machine learning competence accounts for 46.2% of the overall variance in student academic achievement.
  • Skill Breakdown: Teachers scored highest in foundational data literacy (mean = 3.78) but registered their lowest scores in executing predictive pedagogical interventions (mean = 3.02).
  • Primary Influences: Among the specific competencies, the ability to utilize learning analytics exerted the strongest statistical influence on student success ($\beta = 0.534$).
  • Student Performance: The average student report card score stood at 81.6, with 91.1% of the student body successfully surpassing the school's minimum passing standard of 75.

Qualitative records confirmed that educators with high machine learning skills utilized automated tools like Quizizz Analytics, Google Classroom dashboard logs, and Edmodo Premium to evaluate item-by-item student error rates. This allowed them to deploy targeted, digital remedial lessons rather than repeating entire curricula from scratch.

Implications and Real-World Impact

These insights offer vital practical implications for policymakers, school administrators, and global educational institutions. The study establishes that simply purchasing premium digital platform subscriptions is insufficient; schools must actively train educators to interpret data metrics to achieve genuine academic improvements.

By mastering machine learning interfaces, teachers can instantly isolate complex subject matters—such as specific concepts in jurisprudence or grammar—and address them using interactive digital modules. Furthermore, gamified digital assessment methods stimulate student enthusiasm, driving high quiz participation rates and rapid feedback loops that allow struggling students to self-correct quickly.

Expert Commentary

"After every quiz, I first open the Class Summary on Quizizz. I look at which questions students got the most wrong. From there, I create a short video using CapCut and upload it to Google Classroom. So the remediation isn't starting from scratch, but is targeted," stated an Islamic Religious Education teacher identified as G-03, who holds a Master’s degree in Learning Technology and achieved the highest competence ranking in the study.

The research team from Sunan Gresik University emphasizes that while machine learning tools accelerate administrative efficiency, they must serve strictly to complement human mentorship, preserving the essential heart-to-heart connections fundamental to holistic character development.

Author Profiles

  • Sulalatun Nikma is a faculty researcher at Sunan Gresik University with an advanced academic degree in education. Her field of expertise focuses on instructional technology, digital pedagogy, and the optimization of learning environments in secondary education.
  • Ani Khofiati is an academic scholar at Sunan Gresik University specializing in educational curriculum design, technological integration, and data-driven teaching frameworks.
  • Medycha Emhandyaksa is a researcher at Sunan Gresik University whose professional expertise spans educational analytics, quantitative research methods, and modern pedagogical tools.

Source Information

Gloria Florensia Br Surbakti

Diposting oleh Gloria Florensia Br Surbakti

Posting Komentar

0 Komentar