Application of selected theories in this study

This section demonstrates potential capacity of the discussed theories in determining user continuance intention towards M-pesa.

3.2.6.1 PAM

This study adopts three of four constructs of the original PAM (Bhattacherjee, 2001a), and one from the updated PAM (Bhattacherjee et al. 2008). Aforementioned, the original PAM is made up of four constructs (Perceived Usefulness, Disconfirmation, Satisfaction, and Continuance Intention). The researcher selects three of the four constructs and substitutes the perceived usefulness construct for an explicitly labelled construct (post-usage usefulness) to measure user perceived usefulness in a post-adoption era (Bhattacherjee et al. 2008). The revision (post usage usefulness) is particularly important for studies on continuance intention because it helps to forthrightly distinguish it from pre-usage perception of usefulness- ‘perceived usefulness’, and to capture an aggregated perception of usefulness long after initial use. This variable would aid capturing user explicit view of usefulness post- use of the M-pesa service.

Thus, in this study, all the aforementioned constructs of the PAM are adopted for the following reasons: (i) the dependent variable of the PAM directly captures the core objective of this study- to determine user continuance intention towards M-pesa (ii) Post usage usefulness and disconfirmation are validated as critical antecedents of user continuance intention in the literature

(iii) and satisfaction is selected based on extant findings of it being the most potent determinant in the PAM.

3.2.6.2 TTF in this study

Based on reports of the significant value of the TTF in understanding various use cases of technology-enabled product and services, this study adopts two factors (technology-task fit and utilization) from the TTF as potential antecedents and determinants of user continuance intention towards M-pesa. The selection of these two variables and omission of others in the TTF is objective and context driven. One of the RQs in this study is to identify the antecedents of user continuance intention towards M-pesa. The selected two variables from the TTF model are suggested in the literature (e.g., Larsen et al. 2009; Lin, 2012; D’Ambra et al. 2013) as antecedents of user continuance intention because a user who perceives that M-pesa’s characteristics matches his or her task requirement is likely to use the service. Utilization suggests repetitive use of a technology and could therefore determine continuance intention. The other variables in the TTF model (individual characteristics, technology characteristics, and performance impacts) are excluded. The exclusion is executed because individual and technology characteristics are antecedents of TTF, which is not the core focus of the study. Thus, the selection of the TTF serves as a parsimonious and shrewd selection, given suggestions of its influence on utilization which influences continuance (Larsen et al. 2009). Last, performance impact is not included in this study because it is a dependent variable, and the focus of the present study is on user continuance intention and not performance.

3.2.6.3 IS success model in this study

The IS success model is a general model (pre-post) adopted for investigations of technology success. However, given a specific focus of this study on continuance intention- a post-adoption phenomenon, the researcher selects post-adoption applicable constructs from the model for his query. Specifically, the researcher adopts the information quality, system quality and service quality as triad facets of M-pesa attributes. This is because the extant literature on technology characteristics, examined at a general level, are reported to influence behavioural attitudes (e.g., Igbaria, Livari, & Maragahh, 1995; Lim & Benbasat, 2000; Hong, Thong, Wai-Man Wong, 2002) or behavioural outcome (Hsieh, Rai, & Xu, 2011; Setia, Venkatesh, & Joglekar, 2013). The net benefit construct, intention to use, and satisfaction construct are excluded because: (i) net benefit

is a dependent variable and this study does not seek its prediction. (ii) Satisfaction is already adopted, as it is a construct in the PAM (iii) intention to use is a pre-adoption belief and does not feature in a post-adoption era. Lastly, the IS success model has been applied to comprehend behaviour of users of mobile. For example, Chatterjee, Chakraborthy, Sarker, Sarker, & Lau (2009) executed a study to discover potential success drivers in mobile health. Lee and Chung (2009) report that perceptions of system and information quality, and visual quality influence trust and satisfaction with mobile finance. Zhou (2011) adopted the IS success model and complementary factors to assess success drivers of mobile website acceptance. Last, Zhou (2013) examined the impact of the IS success quality factors on continuance intention with technology in China. Collectively, while application of the IS success model is endorsed generally for examining use behaviour, it is infrequently assessed within a developing country perspective.

Moreover, few efforts have been made to test its predictive powers in assessing user behaviour towards mobile money which is a promising technology artefact for the developing regions. Thus, extending a validation test of the model to explain use of mobile money is desirable.

3.2.6.4 Flow theory in this study

Abstractly, flow is multi-faceted, and captures aspects of perceived pleasure, control, and focus (Koufaris, 2002), interest (Ho and Kuo, 2010), and response and consciousness (Guo and Poole, 2009). Extant studies have noted that flow experience is a useful construct for describing human–

computer interactions (Chang & zhu, 2012). Notably, flow experience has been examined in online contexts such as online banking (Lee, Kang, & McKnight, 2007), online games (Lee &

Tsai, 2010), and mobile instant messaging (Zhou & Lu, 2011). These studies submit that flow experience is a salient determinant of user online behaviour. However, scant attention has been devoted to the role of flow experience in emerging technologies such as mobile money continuance intention. An application of the flow concept to this study captures a user’s perception of the basic knowledge and skills that he or she possesses of M-pesa, possible pleasure derived during use of M-pesa, and also the challenges that he or she faces such as: operation difficulty and concerns about security of cash transfers.

3.2.6.5 Trust in this study

Trust is a core factor in the preservation of exchange relations (Blau, 1964). Afore-noted, the concept of trust captures a setup of two actors, where one actor places his trust in the other

(Hardin, 1993, p.507). This scenario is reflected in the mobile money environment, where a user makes a transaction, believing that the M-pesa vendor will deposit, transfer, and permit cash withdrawals as per agreed protocols.

Technology users (M-pesa users) possess finite knowledge and psychical assets, and hence seek to ease the insecurity and intricacies of online transactions by applying cognitive short-cuts (Grabner-Kraeuter, 2002). A valuable psychological alternative is trust, which could be a means to reduce the intricacies of human behaviour in circumstances of incertitude (Luhmann, 1989).

Due to finite influence on the vendor and the absence of verified assurances that a vendor will not act opportunistically; trust becomes a necessary aspect of online-based transactions (Gefen et al.

2003). To define the proposed role of trust in this study, the researcher adopts the conceptualization of trust as applied in an e-commerce context in the extant literature. A study by Gefen et al. (2003) is a preferred starting point, as the scholars conducted a rigorous study on

‘trust’ in on-line shopping, and this work has become one of the most cited articles in trust based studies in the IS literature. Trust as a concept in Gefen et al’s. (2003) study can be said to institutional because in an e-commerce context, a customer would usually make a purchase or transaction without dealing face-to-face with an individual. The customer is said to deal with a legal business entity, hence the term institutional trust. The researcher’s decision to align the present study’s conceptualization of trust with the aforementioned study is due to the like nature of both studies. They both focus on consumer use of an on-line technology platform. As such, trust is applicable to this study because M-pesa is a vendor (entity X), and provides services to user’s (entity Y). Both entity X and Y have an agreed upon way of providing and using services, and this exchange relationship is contingent upon both parties ability to interact as established.

Therefore, M-pesa users have expectations and perceptions of several factors related to an M-pesa service, and M-pesa’s has a duty to deliver on the agreed upon service to its users.