ODLPERF: An Instrument For Measuring Service Quality In An Open And Distance Learning (ODL) Institution

• Main Authors: Ramli Bahroom,, Latifah Abdol Latif,, Ng, Man San
• Format: Conference or Workshop Item
• Published: 2009
• Subjects: LC5201 Education extension. Adult education. Continuing education; LC5800 Distance education
• Online Access: http://library.oum.edu.my/repository/564/1/ODLPERF.pdf; http://library.oum.edu.my/repository/564/

Service quality is important in higher education institutions, particularly in an open and distance learning (ODL) environment, where learner attrition rates are generally higher than that of their conventional counterparts. This has led to a rigorous quest for an effective service quality measurement instrument to provide the institutions with the appropriate information that would help meet the expectations of the students in the provision of quality study programmes, teaching and learning and other educational services. This paper reports on a study that was carried out at Open University Malaysia (OUM) to develop an instrument referred to as ODLPERF for measuring service quality for ODL institutions. The study employed a 29-item Importance-Performance Survey questionnaire that was administered randomly to 2,491 under-graduate learners of OUM in 2008 of which 894 were finally returned and used in this study. Using the AMOS 16.0 computer package and applying the methods of exploratory and confirmatory factor analyses, the study identified four factors, namely tangibles, reliability, assurance and empathy as the key dimensions of service quality. The four-factor model was analyzed based on the maximization likelihood (ML) estimation method. The recommended goodness-of-fit indices of the model were found to be within tolerable ranges (RMSEA =0.06, GFI=0.94, AGFI=0.92, CFI=0.95), suggesting that the model provides a close fit to the data. The scales were also found to show acceptable internal consistency reliability with both Cronbach’s Alpha and construct reliability ranging from 0.8 to 0.9, above the recommended threshold of 0.7. To provide evidence of high convergent validity, the standardized loading estimates for each of the four dimensions were found to be above the recommended threshold of 0.5. The average variance extracted (AVEs) were also found to be within the acceptable range of 0.5-0.6. The discriminant validity for the four dimensions was evident as the AVEs were found to be equal to or greater than the squared correlations between any two dimensions. Collectively, the above tests provide reasonable support for reliability and validity of the service quality construct. This suggests that the proposed ODLPERF is a valid and reliable instrument to measure service quality in an ODL institution. (Abstract from authors)