Security, privacy and institutional logics: conduit to South African citizens' e-government service adoption

Bayaga, Anass

Security, privacy and institutional logics: conduit to South African citizens' e-government service adoption

Thesis / Dissertation

2026

Publisher

University of Cape Town

Department

Department of Information Systems

Faculty

Faculty of Commerce

Abstract

Background and context: This current research examines the acceptance of e-government services in South Africa, advancing privacy calculus theory and incorporating institutional logics to construct a conceptual model for understanding the socio-cultural and technical constructs influencing acceptance. The research assesses the influence of security risks (SR), privacy risks (PR), and institutional logics (InsLog), by examining two research questions: (1) To what extent do security risks, privacy risks, and institutional logics affect e-government acceptance? (2) How can a conceptual model integrate these constructs advance insight of adoption behaviors in developing settings? Methodology: The research hinges on a positivist paradigm, assuming an objective ontological viewpoint where reality exists externally or independently of social constructs, and an epistemological view highlighting empirical testing of causal associations. Data were collected from 149 respondents in urban and peri-urban areas of the City of Umhlathuze, South Africa. The conceptual model was established utilizing partial least squares structural equation modeling (PLS-SEM) to assess direct adoption pathways, interaction effects, and moderated adoption configuration as well as model fit. Theoretical and Conceptual Framework: The conceptual framework incorporates privacy calculus theory and institutional logics to examine how socio-cultural and technical constructs affect e-government acceptance. Institutional logics are postulated to directly affect e-government acceptance by embedding socio-cultural perspectives into respondents decision-making procedures. This adoption configuration investigates the degree to which these norms drive acceptance behaviors. In other words, institutional logic profiles the way citizens view and approach the acceptance of e-government services, rendering them a key determining factor. This is denoted by the pathway InsLog → e-gov adoption. Security risks are projected to negatively influence acceptance by creating perceptions of vulnerability and mistrust in e-government systems. The adoption pathway examines how concerns about security, such as fears of fraud, hacking, or data breaches, can prevent users from accepting these services. This configuration is represented by the pathway SR → e-gov adoption. Analogously, privacy risks are conceived to influence acceptance by influencing citizens' perceptions of data misuse or exposure. This pathway assesses whether concerns about the misuse of personal data or privacy violations considerably reduce the prospect of accepting e-government services. This is represented by the pathway PR → e-gov adoption. Institutional logics are also theorized to moderate the effects of security and privacy risks. This configuration assesses whether socio-cultural norms do reduce or increase the effect of these risks on adoption choices. In this setting, institutional logics may either improve or exacerbate the perceived value of security and privacy concerns, conditional on the cultural and societal context. These connections are characterized by the pathways InsLog x SR → e-gov adoption and InsLog x PR → e-gov adoption. To conclude, demographics (age, education, and income) are assessed for their moderating effects on the connection between institutional logics and e-government adoption. This pathway assesses whether these demographic factors influence how institutional logics affect adoption. These connections are characterized by the pathway InsLog x Demographics → e-gov adoption. This framework combines socio-cultural and technical constructs to provide an inclusive model for understanding acceptance behaviors, especially in developing country settings. By incorporating institutional logics into privacy calculus theory, the framework explores the limitations of existing models, which often overstate technical risks while overlooking the effects of socio-cultural constricts on acceptance. Results: Guided by research question 1: To what extent do security risks, privacy risks, and institutional logics affect e-government acceptance? The findings show that institutional logics significantly affect e-government acceptance (β = 0.444, t = 5.722, p < 0.001), explaining 23.4% of the difference (R² = 0.234). Security risks (β = 0.003, p = 0.951) and privacy risks (β = -0.118, p = 0.130) show no significant direct effect on acceptance, implying their limited influence in this setting. In the case of research question 2: How can a conceptual model integrating these constructs advance understanding of acceptance in developing settings? The conceptual model confirms institutional logics as the leading driver of acceptance, accentuating the role of socio-cultural norms. Moderating effects - demographic factors- (InsLog x Demographics → e-gov adoption, p > 0.05) and the interaction between risks and institutional logics (InsLog x SR → e-gov adoption, β = 0.067, p = 0.314; InsLog x PR → e-gov adoption, β = -0.092, p = 0.284), were non-significant. These results point out the dominance of socio-cultural constructs over technical concerns in e-government acceptance behaviors. Contributions: The research contributes to theory by extending privacy calculus theory to incorporate institutional logics, exploring the gap in understanding the socio-cultural determining factor of e-government acceptance. The conceptual framework focusses on the significant role of institutional logics as a contributing factor of adoption, contrasting the limited effects of security and privacy risks. The model underlines the need for an expansive socio-cultural perception in assessing e-government acceptance, specifically in developing settings were institutional norms model user behaviors. This theoretical development offers a basis for incorporating cultural and demographic nuances into technology adoption models. The research establishes the value of PLS-SEM in evaluating multi-faceted adoption configurations and validating a conceptual model that incorporates socio-cultural construct into e-government acceptance research. This methodological consideration improves the tools accessible for examining technology adoption in diverse settings. Recommendations: The results offer actionable direction for policymakers and practitioners: • Design e-government programs that align with socio-cultural norms and institutional logics to improve citizens commitment and engagements. • Execute targeted digital literacy programs to reduce inequalities in access and usability. To conclude, this research advances both theoretical and methodological insights of e-government acceptance by offering a validated conceptual framework that highlights the predominance of institutional logics. These understandings serve as a ground for developing comprehensive, culturally supported digital public services in developing settings.