Predicting user continuance intention towards M-pesa in Kenya

摘要

Purpose - The purpose of this paper is to integrate established information systems theories (post-acceptance model (PAM) and task-technology fit (TTF)) to model and test determinants of user continuance intention within Kenya toward a proliferated mobile money service called M-pesa. Design/methodology/approach - A positivist method informed the design of the study. A survey instrument was developed and administered amongst M-pesa users in Kenya, yielding 618 responses. Selection of users followed a systematic sampling technique. Afterwards, structural equation modelling (SEM)-partial least squares ("SEM-PLS") was used to examine the measurement and structural model of the study. Findings - The results revealed an unexpected finding that TTF's utilization has a stronger path coefficient than PAM's satisfaction in predicting user continuance intention toward M-pesa. This is contrary to most extant literature reports that satisfaction is the dominant predictor of usage continuance. Also, unexpected, the results revealed a lack of significant influence between the PAM's post-usage usefulness (PUU) and satisfaction. Again, the relationship between PUU and satisfaction in most extant studies is reported as significant. Originality/value - The study suggests that saliency order of determinants of user continuance intention could differ within the developing world and developed world.