DOI: 10.12688/f1000research.187984.2 ISSN: 2046-1402

Ceiling Effects and Item Discrimination in Satisfaction-Based Instructional Facilitation Measures Among Indonesian Pre-Service Science Teachers

Febrian Virijai, Eka Cahya Prima, Fitri Khoerunnisa, Rika Rafikah Agustin, Dens E.S.I. Asbanu, Resti Warliani, Hermansyah Hermansyah, Andreas Suparman
Show PDF Cite
Background Satisfaction-based surveys are widely used in educational quality assurance, yet their ability to distinguish among respondents may be limited when responses are concentrated at the upper end of the scale. This study examined ceiling concentration and item discrimination in instructional facilitation measures among Indonesian pre-service science teachers, together with their observed associations with learning motivation. Methods A cross-sectional measurement diagnostic analysis was conducted using responses from 348 pre-service science teachers from 29 higher education institutions in Indonesia. The analysis included 22 instructional facilitation items covering Lecture Implementation, Teaching Material Use, and Science Practicum, together with six Learning Motivation items. Ceiling concentration was assessed using a 15% maximum-response benchmark, while corrected item-total correlations were examined as indicators of item discrimination. Facilitation-Motivation associations were evaluated using 132 Pearson correlations with Benjamini-Hochberg false discovery rate correction. Instructional facilitation scores were also compared across study programme, semester, gender, age, and academic achievement groups using Kruskal-Wallis tests with epsilon-squared effect sizes. Results All 22 instructional facilitation items exceeded the 15% ceiling benchmark, with approximately half of respondents selecting the maximum response category. Corrected item-total correlations were positive but remained below the commonly used r = .30 item-discrimination benchmark. Facilitation-Motivation associations were consistently small, and none of the 132 item-pair correlations remained statistically significant after false discovery rate correction. Demographic comparisons also showed little variation in instructional facilitation scores, with small effect sizes across the groups examined. Conclusions The findings indicate substantial ceiling concentration and limited item discrimination in the instructional facilitation measures examined in this sample. These patterns co-occurred with weak observed associations with learning motivation but do not establish that ceiling effects caused or attenuated those associations. The study highlights the importance of evaluating response distributions and item discrimination before satisfaction-based measures are used in relational or predictive analyses.

More from our Archive