Graphical diagnostics in proportional odds models—Empirical study on determinants of FinTech payment service diffusion by SMEs in Italy

FinTech, the merging of finance and modern Internet-based technology, has rapidly presented itself as a disruptor to traditional business sector. In this paper, we examine the determinants of the use of FinTech payment services by Italian SMEs, using a conceptual framework based on the technology diffusion theories. In this study we consider the proportional odds model for ordinal logistic regression and we consider different approaches to assessing the goodness of fit of the model. Model diagnostic is an essential element in statistical modeling of business data, since, it helps researcher to re-evaluate their working models in order to inform business strategies. To test the fit of the model and check the assumptions in the ordinal regression model (i.e., misspecification mean structure, proportional, heteroscedasticity, etc.) diagnostic plots based on surrogate residuals are shown. The findings states that the innovation-firm level, firm’size, and a gendered governance positively impact on the Fintech payment services diffusion. The reported findings per the study are strong, robust, and reliable since the various proposed models employed in the study are significantly fit and valid.