Integrating Technology Acceptance, Agency, and Systems Theories to Explain Digital Accounting's Impact on Financial Reporting Quality: A Mixed-Methods Study of Nigerian SMEs

Abstract

Purpose: This study integrates the Technology Acceptance Model (TAM), Agency Theory, and Systems Theory to examine how digital accounting system adoption relates to financial reporting quality in small and medium-sized enterprises (SMEs), with internal control effectiveness specified as a mediating mechanism. The study also develops three theory-informed propositions from the combined quantitative and qualitative evidence. Design/methodology/approach: An explanatory sequential mixed-methods design was used. Quantitative data were obtained from 207 SME respondents in Lagos State, Nigeria. Of 480 questionnaires targeted, 230 were returned and 23 were excluded because of incompleteness or non-qualification, leaving 207 usable responses and an adjusted response rate of 43.1%. Mediation was examined using Hayes' PROCESS Macro Model 4 with 5,000 bootstrap resamples. Thirteen survey participants were subsequently interviewed purposively, and the interviews were analysed thematically using Braun and Clarke's six-phase approach. Findings: The quantitative analysis found no significant total effect of digital accounting system adoption on financial reporting quality (B = 0.031, p = 0.673) and no significant effect on internal control effectiveness (B = 0.029, p = 0.683). Internal control effectiveness was negatively associated with self-reported financial reporting quality (B = −0.159, p = 0.028). The full-sample indirect effect was not significant (B = −0.0046, 95% CI [−0.0334, 0.0155]). An exploratory comparison among digital-system users found 68 basic-use respondents who reported Excel use and 118 more-extensive-use respondents who reported dedicated accounting software. The latter group reported higher ICE and FRQ, and the exploratory mediation effect was significant (indirect effect = 0.084, 95% CI [0.021, 0.157]). This subgroup analysis was not a formal test of moderated mediation and was not treated as confirmatory evidence. Qualitative findings consistently described audit trails, segregation of duties, approval workflows, and other controls as mechanisms through which participants perceived reporting improvements. Originality/value: The study does not claim that the three established theories are empirically extended by the present data. Instead, it develops three provisional refinements for future testing: distinguishing operational usefulness from control usefulness in TAM; treating the negative ICE - FRQ association as a possible critical self-assessment or measurement phenomenon rather than established 'suppressed reporting'; and specifying affordance actualisation as a possible condition for productive interaction between technology and control subsystems.