[论文解读] Measuring User Perceived Security of Mobile Banking Applications

Richard Apaua, Harjinder Singh Lallie|arXiv (Cornell University)|Jan 9, 2022

Technology Adoption and User Behaviour被引用 11

一句话总结

本研究通过在315名参与者中应用基于协方差的结构方程模型（CB-SEM）的经验证扩展UTAUT2模型，实证测量了移动银行应用中的用户感知安全。研究发现，感知安全、机构信任和技术信任显著预测采纳意愿和使用行为，且这些影响受到年龄、性别、经验、收入和教育程度的调节——解释了79%的行为意图方差。

ABSTRACT

Mobile banking applications have gained popularity and have significantly revolutionised the banking industry. Despite the convenience offered by M-Banking Apps, users are often distrustful of the security of the applications due to an increasing trend of cyber security compromises, cyber-attacks, and data breaches. Considering the upsurge in cyber security vulnerabilities of M-Banking Apps and the paucity of research in this domain, this study was conducted to empirically measure user-perceived security of M-Banking Apps. A total of 315 responses from study participants were analysed using covariance-based structural equation modelling (CB-SEM). The results indicated that most constructs of the baseline Extended Unified Theory of Acceptance and Use of Technology (UTAUT2) structure were validated. Perceived security, institutional trust and technology trust were confirmed as factors that affect user's intention to adopt and use M-Banking Apps. However, perceived risk was not confirmed as a significant predictor. The current study further revealed that in the context of M-Banking Apps, the effects of security and trust are complex. The impact of perceived security and institutional trust on behavioural intention was moderated by age, gender, experience, income, and education, while perceived security on use behaviour was moderated by age, gender, and experience. The effect of technology trust on behavioural intention was moderated by age, education, and experience. Overall, the proposed conceptual model achieved acceptable fit and explained 79% of the variance in behavioural intention and 54.7% in use behaviour of M-Banking Apps, higher than that obtained in the original UTAUT2. The guarantee of enhanced security, advanced privacy mechanisms and trust should be considered paramount in future strategies aimed at promoting M-Banking Apps adoption and use.

研究动机与目标

• 在网络威胁日益增加的背景下，实证测量用户在移动银行应用中的感知安全。
• 检验感知安全、信任和风险对用户采纳和使用意愿的影响。
• 探究人口统计因素（年龄、性别、经验、收入、教育程度）对感知安全、信任与行为意图之间关系的调节作用。
• 在移动银行安全感知背景下，验证并扩展UTAUT2模型。
• 为通过增强安全与信任机制提升移动银行采纳率，提供可操作的见解。

提出的方法

• 采用经验证的扩展UTAUT2概念模型，评估移动银行应用中用户的感知安全。
• 从315名参与者收集调查数据，测量感知安全、机构信任、技术信任、感知风险和行为意图等构念。
• 应用基于协方差的结构方程模型（CB-SEM）检验测量模型和结构模型的拟合度。
• 通过多组分析评估年龄、性别、经验、收入和教育程度对关键关系的调节效应。
• 使用CFI、TLI、SRMR和RMSEA评估模型拟合度，并评估行为意图和使用行为的决定系数（R²）。
• 通过验证性因子分析验证测量模型，并评估收敛效度与区别效度。

实验结果

研究问题

• RQ1 感知安全、机构信任和技术信任在多大程度上能预测用户采纳和使用移动银行应用的意愿？
• RQ2 在移动银行应用背景下，感知风险是否是行为意图的显著预测因子？
• RQ3 人口统计因素（年龄、性别、经验、收入、教育程度）如何调节感知安全与行为意图之间关系？
• RQ4 人口统计因素如何调节技术信任对行为意图的影响？
• RQ5 与原始UTAUT2模型相比，所提出的模型在解释行为意图和使用行为方差方面有多大程度的提升？

主要发现

• 感知安全、机构信任和技术信任被证实是用户采纳和使用移动银行应用意愿的显著预测因子。
• 感知风险并非行为意图的显著预测因子，与初始预期相反。
• 感知安全和机构信任对行为意图的影响显著受到年龄、性别、经验、收入和教育程度的调节。
• 感知安全对实际使用行为的影响受到年龄、性别和经验的调节。
• 技术信任对行为意图的影响受到年龄、教育程度和经验的调节。
• 所提出的模型解释了79%的行为意图方差和54.7%的使用行为方差——高于原始UTAUT2模型。