Configural measurement equivalence testing of the comprehensive trait based model of self-regulated learning in engineering undergraduates
By: Chakraborty, R
.
Contributor(s): Chechi, V. K.
Publisher: Sangli K.E. Society's Rajarambapu Institute of Technology 2024Edition: Vol.37(3), Jan.Description: 143-156p.Subject(s): Humanities and Applied SciencesOnline resources: Click here
In:
Journal of engineering education transformations (JEET)Summary: The present study extended the integrative trait model of self regulated learning by [30] by including the remaining behavioral and emotional components. Measurement invariance testing was conducted to check for the equivalence of the model with respect to gender, stream and batch among 488 (351 male and 137 female; 321 Computer Science and 167 Mechanical; 263 IInd Year and 225 IIIrd Year) engineering students selected through stratified random sampling from the three regions, Majha, Doaba and Malwa, of the Punjab state of India. [25] measurement invariance criterion of ΔCFI be less than or equal to 0.01 was used to test for equivalence of the model across the selected groups, since it is unaffected by model complexity, sample size and unrelated to overall fit measures. The revised integrative trait model of self regulated learning among engineering undergraduates was found to be Configural measurement invariant with respect to gender, stream and batch using SPSS Amos Ver. 23.0, meaning that the construct of self regulated learning is conceptualized by the participants across the groups of the study in same way. Reliability of the instrument's 14 variables was estimated using Pearson's correlation-based Cronbach's alpha (using SPSS Statistics Ver. 23.0) and Polychoric correlation based ordinal alpha (using R Package Psych), along with estimation of attenuation index to show the extent of underestimation of the vital psychometric property by assuming the data of Likert scale based questionnaires as continuous interval and on ignoring its categorical ordinal nature. The effect size of the validated model conducted as part of post-hoc power analysis using semPower R package was found to be satisfactorily high at 0.941. The academic implications of the study with respect to engineering education in the country are discussed.
| Item type | Current location | Call number | Status | Date due | Barcode | Item holds |
|---|---|---|---|---|---|---|
Articles Abstract Database
|
School of Engineering & Technology Archieval Section | Not for loan | 2024-1597 |
The present study extended the integrative trait model of self regulated learning by [30] by including the remaining behavioral and emotional components. Measurement invariance testing was conducted to check for the equivalence of the model with respect to gender, stream and batch among 488 (351 male and 137 female; 321 Computer Science and 167 Mechanical; 263 IInd Year and 225 IIIrd Year) engineering students selected through stratified random sampling from the three regions, Majha, Doaba and Malwa, of the Punjab state of India. [25] measurement invariance criterion of ΔCFI be less than or equal to 0.01 was used to test for equivalence of the model across the selected groups, since it is unaffected by model complexity, sample size and unrelated to overall fit measures. The revised integrative trait model of self regulated learning among engineering undergraduates was found to be Configural measurement invariant with respect to gender, stream and batch using SPSS Amos Ver. 23.0, meaning that the construct of self regulated learning is conceptualized by the participants across the groups of the study in same way. Reliability of the instrument's 14 variables was estimated using Pearson's correlation-based Cronbach's alpha (using SPSS Statistics Ver. 23.0) and Polychoric correlation based ordinal alpha (using R Package Psych), along with estimation of attenuation index to show the extent of underestimation of the vital psychometric property by assuming the data of Likert scale based questionnaires as continuous interval and on ignoring its categorical ordinal nature. The effect size of the validated model conducted as part of post-hoc power analysis using semPower R package was found to be satisfactorily high at 0.941. The academic implications of the study with respect to engineering education in the country are discussed.
Click on an image to view it in the image viewer
There are no comments for this item.