Structural variable validation of an Online Learning Response Behavior (OLRB) instrument: A comparison analysis of three extraction methods of Exploratory Factor Analysis

• Published in: Asian Association of Open Universities Journal
• Main Author: Azman Ong M.H.; Mohd Yasin N.; Ibrahim N.S.
• Format: Article
• Language: English
• Published: Emerald Publishing 2022
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85139242719&doi=10.1108%2fAAOUJ-04-2022-0054&partnerID=40&md5=0b2dfccc43e7712a41f9d1c1559d65e3

Purpose: Measuring internal response of online learning is seen as fundamental to absorptive capacity which stimulates knowledge assimilation. However, the evaluation of practice and research of validated instruments that could effectively measure online learning response behavior is limited. Thus, in this study, a new instrument was designed based on literature to determine the structural variables that exist in the online learning response behavior. Design/methodology/approach: A structured survey was designed and distributed to 410 Malaysian students enrolled in higher-education institutions. The questionnaire has 38 items, all of which were scored using a seven-point likert scale. To begin with, exploratory factor analysis with three types of extraction methods (i.e. principal component, principal axis factoring and maximum likelihood) was used as the method for comparing the outcomes of each extraction method's grouping variables by constantly using a varimax rotation method. In the second phase, reliability analysis was performed to determine the reliability level of the grouping variables, and finally, correlation analysis was performed to determine the discriminant nomological validity of the grouping variables. Findings: The findings revealed that nine grouping variables were retrieved, with all items having a good value of factor loading and communalities, as well as an adequate degree of reliability. These extracted variables have good discriminant and nomological validity, as evidenced by correlation analysis, which confirmed that the directions of relationships among the extracted dimensions follow the expected theory (i.e. positive direction) and the correlation coefficient is less than 0.70. Research limitations/implications: This study proposes a comprehensive set of questionnaires that measure the student's online learning response behavior. These questionnaires have been developed on the basis of an extensive literature review and have undergone a rigorous process of validity and reliability for the purpose of enhancing students' online learning response behavior. Originality/value: This study's findings will aid academic practitioners in assessing the online learning response behavior of students, as well as enhancing the questionnaire's boost factor when administered in an online learning environment. © 2022, Mohd Hanafi Azman Ong, Norazlina Mohd Yasin and Nur Syafikah Ibrahim.