On The Formation of End-User Computing Satisfaction: A Proposed Model And Measurement Instrument

1 INTRODUCTION The measurement of end-user computing satisfaction has had a long history within the IS discipline. To date, a number of attempts have been made to capture the overall post hoc evaluation that end users have regarding the use of an IS system along with the most immediate antecedent factors that form this satisfaction (Doll The purpose of this paper is to build further upon the existing body of research by providing a new conceptual perspective on how end-user computing satisfaction (EUCS) is formed and how it ought to be measured. In addition, this paper operationalizes this new perspective by providing a new measurement instrument for empirical testing (see Appendix). As a baseline for comparison, we use Doll and Torkzadeh's EUCS model (Doll and Torkzadeh, 1991) since this is probably one of the most well-known and frequently tested model in the literature. We also develop additional measures for the 5 constructs in the model. This is considered necessary in order to avoid methodological problems dealing with actual or empirical under-identification when 2 used in a structural equation modelling analysis. 2 The extent to which our new approach provides greater explanatory power over the baseline model is considered. In order to discuss our model of End-User Computing Satisfaction, we start with a definition of end-user satisfaction. We define end-user satisfaction with an information system as the overall affective evaluation an end-user has regarding their experience related with the information system. The term " experience " can be made more specific to focus upon different aspects related to the information system (e.g., computing, training, etc.). In general, past studies have focused primarily on the satisfaction measurement of the computing/use aspect of a system. But it may well be satisfaction with other activities other than system use (e.g., training, participation or involvement in development or selection) that may also be of value in predicting subsequent behaviour (e.g. utilisation) or performance. Alternatively, a global measure of the end-user computing experience may equally be of value. For the purposes of this study and to be consistent with prior research, we will focus on the system use. But we note that our instrument is meant to be a general one that a researcher can easily modify to include other aspects related to End-User Satisfaction. Methodologically, we recommend that the explicit use of the term satisfaction or a synonymous term be applied in either Likert or …