SERVQUAL in an internal nonprofit market : psychometric issues

Abstract

Quality of service, as perceived by the customer, has been shown by research to be a critical factor contributing to organizational performance in recent years. Therefore, the management of service quality is a key variable, and for service quality to be managed, it has to be measured. There have been significant advances in the measurement of service quality in the past fifteen years, resulting in a stream of research, mostly concentrating on the external customers of profit-seeking firms. A key factor driving this research was the development of an apparently reliable, valid instrument for the measurement of service quality. This instrument called SERVQUAL, was developed by US researchers A. "Parsu" Parasuraman, Valarie Zeithaml and Len Berry. It has spawned an enormous debate in the marketing literature, leading to the further exploration and refinement of the dimensions of the service quality construct. While the use of SERVQUAL has been extensively investigated in external markets, and in for-profit firms, less attention has been given to its use, and more importantly, its reliability and validity in internal markets, and in not-for-profit organizations. These settings are becoming increasingly important from a services marketing perspective. Internal markets (where fellow employees are also customers) are being subjected to market testing, and many services previously provided within the organization are being outsourced. In order to survive, many functions such as information systems, training, catering and cleaning are being forced to market their services internally, and this includes assessing service quality, and improving it. Likewise, private and public nonprofit organizations are coming under increasing scrutiny, as donors and taxpayers alike become evermore concerned about the value gained from the expenditures made by these organizations with their funds. In this study, the SERVQUAL instrument was used to measure service quality as perceived by the internal customers of a large IT department within an extensive government organization. The main objectives of the study were to assess the psychometric properties of the SERVQUAL instrument in this setting. It was found that SERVQUAL generally performs well under these circumstances, with regard to reliability, construct, convergent and nomological validity. However, the instrument appears to be problematical in terms of discriminant validity. This is probably less attributable to the measurement situation as to the instrument itself, for the finding mirrors evidence from the literature. The study also identifies implications for management, and opportunities for future research.