Trust in AI: scale development and validation for online distance learners

Area of Science:

• Educational Technology
• Artificial Intelligence in Education
• Psychometrics

Background:

• Student trust in artificial intelligence (AI) is crucial for AI-supported learning environments.
• Existing instruments for measuring student trust in AI in education are limited.
• A validated scale is needed to assess trust in AI within online distance education.

Purpose of the Study:

• To develop and validate a multidimensional scale for measuring student trust in AI systems in online distance education.
• To provide a reliable instrument for research on student-AI interactions in higher education.

Main Methods:

• Sequential multi-stage scale development: literature review, expert evaluation, exploratory factor analysis (EFA), and confirmatory factor analysis (CFA).
• Data collected from 837 distance learning students.
• Psychometric properties assessed, including internal consistency, model fit, and measurement invariance.

Main Results:

• A 21-item, five-factor scale was developed, explaining 63.20% of the variance.
• CFA confirmed good model fit (χ²/df = 2.17, CFI = .95, TLI = .94, RMSEA = .06).
• The scale demonstrated high internal consistency (α = .94, ω = .94) and metric invariance across gender.

Conclusions:

• The developed scale is a reliable instrument for assessing student trust in AI in online learning.
• Trust in AI is a distinct perception influencing student engagement with AI learning tools.
• Findings have practical implications for designing and implementing AI-supported learning in higher education.