The aim of this study is to examine how to maintain users of the Allo Bank mobile banking application through continuance intention, the UTAUT model, and the UATC variables. This research is classified as a quantitative study, with data collection conducted via a survey involving 310 respondents and 51 questions. However, 8 respondent data points were eliminated due to identified outliers, in order to meet the multivariate normality data requirements. The data analysis technique used is the Covariance-based Structural Equation Model (CB-SEM), with the Maximum Likelihood (ML) approach for parameter estimation, relying on multivariate normal distribution through AMOS software. When the multivariate normality test using skewness-kurtosis did not meet the standard of < 2.58, Mahalanobis D-Squared was employed to identify outliers and remove data points with Mahalanobis values > 87.97, resulting in the elimination of 8 data points. Reliability testing was performed using Cronbach’s Alpha, validity was assessed through standardized regression weight, and model fit was evaluated using RMSEA, CFI, and Chi-Square tests. The statistical test results indicate that the reliability test exceeded 0.9, meaning its reliability was good. The multivariate normality test produced a value of 2.245, indicating acceptable conditions. The model fit tests showed a goodness of fit based on RMSEA (0.047), CFI (0.928), and Chi-Square (1.673). Additionally, the path coefficient test results were consistent with the theoretical expectations for each variable, including Performance Expectancy, Effort Expectancy, Social Influence, Facilitating Conditions, Unfavorable Attitude toward Cash Payment, and Satisfaction, all of which had significant effects on Continuance Intention. From a technical perspective, the Allo Bank mobile banking application can provide a reliable system for users, as Performance Expectancy and Effort Expectancy significantly influence Continuance Intention. The study also suggests that the application can enhance the social influence of users to encourage application usage, as indicated by the social influence variable. Additionally, users displayed an unfavorable attitude towards cash transactions based on UATC respondent feedback.
Read full abstract