The development of a situational smartphone addiction scale for students in the Faculty of Education using Multidimensional Item Response Theory

Authors

  • Manaathar Tulmethakaan Department of Educational Measurement and Research, Faculty of Education, Srinakharinwirot University, Bangkok 10110, Thailand
  • Orn-uma Charoensuk Department of Educational Measurement and Research, Faculty of Education, Srinakharinwirot University, Bangkok 10110, Thailand
  • Patcharaporn Srisawat Department of Educational Psychology and Guidance, Faculty of Education, Srinakharinwirot University, Bangkok 10110, Thailand

DOI:

https://doi.org/10.34044/j.kjss.2026.47.4.15

Abstract

This study aimed to develop a situational test for smartphone addiction among
students in the Faculty of Education using Multidimensional Item Response
Theory (MIRT). Qualitative data were gathered through interviews with 18
students at risk of smartphone addiction. Insights from these interviews were used
to construct situational items that reflect realistic challenges aligned with the daily
lives, developmental stage, and decision-making context of late adolescents. These
scenarios were then used to collect quantitative data from 2,186 students across
four regions of Thailand. The test’s psychometric quality was examined using
Classical Test Theory (CTT), while structural validity was assessed via Item
Response Theory using ConQuest 2.0. The findings revealed that most at-risk
students lacked control over smartphone use, leading to interpersonal issues, health

concerns, and reduced academic performance. The finalized test comprises 13 age-
appropriate situational scenarios covering three components: Overuse, Daily-Life

Disturbance, and Withdrawal. It employs a 4-point rating scale (0–3), with
discrimination indices ranging from 0.366 to 0.614 and a reliability coefficient of
0.828. Internal construct validity assessment indicated that the multidimensional
model provided a significantly better fit than the unidimensional one ( χ2 =
161.6025, df = 5, p < .05). Weighted MNSQ values ranged from 0.75 to 1.33, and
no disordered thresholds were found. The use of MIRT enabled more accurate
analysis of students’ responses across multiple behavioral dimensions. These
findings support the scale’s structural validity and the relevance of using situational
questions tailored to the context and developmental stage of the target population.

Downloads

Published

13-11-2026

How to Cite

Tulmethakaan, M., Charoensuk, O.- uma, & Srisawat, P. (2026). The development of a situational smartphone addiction scale for students in the Faculty of Education using Multidimensional Item Response Theory. Kasetsart Journal of Social Sciences, 47(4), 470415. https://doi.org/10.34044/j.kjss.2026.47.4.15

Issue

Section

Research articles

Categories