Applying Atride Sterling's content analysis to analyze FinTech business model indicators for corporate social responsibility practices and improving the company's financial performance

Abstract

The present study aimed to investigate the indicators of the FinTech business model in alignment with corporate social responsibility (CSR) practices and the improvement of corporate financial performance. The participants of this research included university professors and managers. Individuals were selected using a purposive sampling method. The sample consisted of 20 experts and specialists. The data collection instrument comprised two components: (1) the examination and analysis of upstream documents and financial planning documents in the library section, and (2) semi-structured interviews in the field section. The semi-structured interviews with participants continued until the point of theoretical saturation. For the analysis of qualitative data, the thematic analysis method based on the Attride-Stirling model was employed. To ensure validity, the interview questions were reviewed and approved by three financial planning and management experts—one holding a master’s degree and two holding PhDs. To assess reliability, Krippendorff’s alpha coefficient was used, which was confirmed. The ATLAS.ti software was utilized for the thematic analysis. The results from the factor analysis indicated that among the 192 existing indicators (items), 48 first-order constructive themes were identified, leading to the derivation of 12 second-order constructive theme categories. The second-order constructive themes of the research model included: service innovation, operational financial transparency, organizational social accountability, environmental sustainability, customer-centric value creation, digital data security, sustainable profitability growth, equity in access, intelligent risk management, stakeholder commitment, organizational decision-making agility, and alignment with social mission. Ultimately, four quantitative criteria were used to evaluate credibility, transferability, confirmability, and dependability. The results were as follows: the level of expert agreement, measured using Holsti’s coefficient (PAO) or “Percentage Agreement Observed,” was found to be 0.810, which is a significant value. Given the critiques of the Holsti method, the P-Scott index was also calculated, yielding a value of 0.813. The fourth index used to estimate qualitative research credibility was Cohen’s Kappa coefficient, which reached 0.804 in this study. Finally, Krippendorff’s alpha was also used, and it was estimated at 0.852 in this study.