NETWORK PSYCHOMETRICS BASED VALIDATION OF ACADEMIC EMOTIONAL REGULATION QUESTIONNAIRE (AERQ)

Abstract

The academic emotional regulation questionnaire (AERQ), developed by Buric et al. (2016), measures
the trait based emotional regulation ability in academics in University students. The present study applied
network psychometrics analysis to validate the scale in the Indian context. 496 students from the School
of Mechanical engineering, School of Computer Science engineering and the School of Hotel
Management and Tourism of Lovely Professional University, Phagwara, Punjab, India, were the sample
subjects. Mahalanobis distance was computed to detect outliers, removal of which lead to the final
sample size of the study being 443. Exploratory graph analysis (EGA) using the package eganet in
R/RStudio, lead to the extraction of the original eight dimensions of the scale, with one item from social
support dimension showing anomaly. As per the rules of (EGA), it was removed the test was re-run to
obtain the graph of the network. The package lavaan was used to conduct confirmatory factor analysis
using the estimator “WLSMV” for ordinal data, and robust CFI, TLI, RMSEA and SRMR were 0.962,
0.958, 0.026 and 0.048, well below the desired benchmarks, indicating excellent fit. Under network
inference analysis, the state-of-the-art edge weight accuracy using non-parametric bootstrap confidence
interval, correlation stability coefficient and bootstrapped difference tests were conducted to study the
significance of the obtained estimates using the package bootnet. Structural consistency is computed
under reliability analysis instead of internal consistency using the packages bootega, eganet and psych.
The plots are generated using the package qgraph. The significance of this research exercise in the
Indian context is discussed.