Structural equation modeling in the acceptance of internet banking in the city of Split

Abstract

Electronic banking and electronic commerce have lately become an inevitable aspect of financial services. According to data about the percentage of users of Internet banking in Europe, Croatia is at the bottom of the list, with a low number of users. Thus, the question of the acceptance and use of this kind of technology arises. This research uses structural equation modeling to explore whether the motivation for the acceptance and use of Internet banking can be explained by the technology acceptance model (TAM) in the city of Split, Croatia. TAM explains the intent of using information systems through the perceived ease of use and the perceived usefulness of a system. It consists of multiple causal relationships, so structural equation modeling is adequate for hypothesis testing. A survey analysis was designed and applied on a sample of 282 working residents of Split. To make the results more credible, the gender and age structure of the sample was harmonized with the population. It is concluded that both elements of TAM positively influence the acceptance of Internet banking in Split. Since there is not much research on the use of Internet banking in Croatia and Southeast Europe, this research contributes to the poor amount of literature in this scientific area for these countries. The research findings can also help banks to understand the factors of Internet banking acceptance and to develop expansion strategies. Future research could include a crosscountry comparison of individuals' perception about Internet banking, as well as the dependence relationships of TAM factors.

Does financial behavior mediate the relationship between self-control and financial security?

“…Cronbach's alpha coefficients were used for the reliability analysis. Acceptable values for these coefficients are mostly in the range from 0.70 to 0.95, whereas some researchers consider the value of 0.60 adequate [27]. Cronbach's alpha coefficients for all constructs are above the threshold of 0.7, indicating an internal consistency of the items.…”

“…Testing for discriminant validity was conducted according to the Fornell and Larcker criterion. Furthermore, reliability analysis was used to "measure the convergence among a set of items representing a latent construct and to define the extent to which the variables are consistent in their values" [8,27,29].…”