Psychometric Properties of the Online Learner Satisfaction Instrument: Construct and Convergent Validity, and Measurement Invariance

• In Event: AERA Roundtable Session Friday 9:50 am Four Seasons Ballroom 2-3
  In Roundtable Session: Challenges and Opportunities in Measurement and Assessment in Higher Education (Table 2)

Abstract

The present study evaluated the validity of the Online Learner Satisfaction Instrument (OLSI) in terms of construct and convergent validity, and measurement invariance across demographic groups. The study results revealed that the factor structures generally confirmed to the original factor structure proposed in the instrument. Also, some subfactors showed a high correlation, suggesting a potential unidimensionality within each factor. The results of the convergent validity showed that the course subscale had high correlations with learner and instructor subscales, yielding inter-subscale correlations. The results of the measurement invariance testing showed that OLSI was interpreted the same way across participants of different gender and ethnic groups. However, measurement invariance was not supported across different educational levels (undergraduate vs. graduate).

Authors

• Stella Yun Kim, University of North Carolina - Charlotte

• Doris U. Bolliger, Texas Tech University

• Florence Martin, North Carolina State University