Implementation Success Model in Government Agencies: A Case of a Centralized Identification System at NASA

Abstract

A recent presidential directive mandated that all U.S. government agencies establish a centralized identification system. This study investigated the impact of users’ involvement, resistance, and computer self-efficacy on the implementation success of a centralized identification system. Information System (IS) usage was the construct employed to measure IS implementation success. A survey instrument was developed based on existing measures from key IS literature. The results of this study indicated a strong reliability for the measures of all constructs (user involvement, computer self-efficacy, user’s resistance, and IS usage). Factor analysis was conducted using Principal Component Analysis (PCA) with Varimax rotation. Results of the PCA indicate that items of the constructs measured had high validity, while Cronbach’s Alpha for each factor demonstrates high reliability for all constructs measured. Additionally, results of a structural equations modeling analysis using Partial Least Square (PLS) indicate that computer self-efficacy and user involvement had positive significant impact on the implementation success. However, the results also demonstrated that user’s resistance had no significant impact on IS usage, while end user involvement had a strong negative impact on user’s resistance.