the measurement invariance of job diagnostic survey (jds) across three university student groups

Purpose: The main purpose of this study is to apply a multigroup confirmatory analysis to examine the measurement invariance (MI) of the adapted version of the Job Diagnosis Survey (JDS) as a measurement tool that analyses the relationship between the features of teaching methodologies with university students’ motivation and satisfaction across data collected on different degrees and academic years.

Design/methodology/approach: Confirmatory factor analysis was carried out using a multigroup structural equation model, using the program EQS 6.1 to test the invariance of the adapted version of JDS in a sample constituted by 535 student of a Spanish public university. The assessment of invariance included the levels of configural, metric, scalar, covariance and latent variables invariance. Several goodness-of-fit measures were assessed.

Findings: The results show that measurements are equivalent at the configural, metric, covariance and latent factors invariance. Although the hypotheses of scalar invariance is rejected, results suggest that JDS is partial strict invariant and has satisfactory psychometric properties on all samples.

Research limitations/implications: The sample is framed in university students aged between 18 and 30 and for a questionnaire on teaching methodology and students' satisfaction in the context of a Spanish university and the generalization to other questionnaire, or population, should be proved with specific data. Furthermore, the sample size is rather small.

Originality/value: In the current process of change that is taking place in universities according to the plan developed by the European Space of Higher Education, focused on increasing the student skills, validate instruments as the satisfaction scale of JDS, are necessary to evaluate students’ satisfaction with new active methodologies. These findings are useful for researchers since they add the first sample in which the MI of a student’s satisfaction survey is tested.