AI and online purchase decisions: The mediating role of attitude

  • 11 Views
  • 1 Downloads

Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 International License

Type of the article: Research Article

Abstract
By altering how people evaluate information, utilize technology, and make purchasing decisions, Artificial Intelligence (AI) is transforming the way customers shop online. This study aims to examine the mediating role of attitudes toward AI in the relationship between cognitive and psychological factors and online purchase decisions. A mixed-methods design with two phases was used. To improve and culturally validate the measurement items, five marketing and e-commerce specialists were interviewed in May 2025 as part of the qualitative phase of the study. During the quantitative phase, customers in Ho Chi Minh City who regularly shop on major e-commerce platforms were invited to complete a structured online survey between June 1 and July 15, 2025. A self-administered online questionnaire disseminated via Facebook and Zalo yielded 458 valid responses. Respondents were selected because they frequently interact with AI-enabled features, such as recommendation systems, chatbots, and personalized interfaces, making them highly relevant to the study’s objectives. SmartPLS 4.0 and SPSS 26.0 were used to analyze the data. Cronbach’s Alpha, EFA, CFA, and PLS-SEM were used to confirm reliability, convergent validity, and discriminant validity. The results indicate that Attitude toward AI are significantly influenced by Perceived Usefulness (PU), Ease of Use, Trust, and Enjoyment. The desire to make an online purchase is strongly positively impacted by attitude (β = 0.327; p < 0.001; R2 = 0.135). These findings emphasize the importance of enhancing trust, usability, and emotional value in AI-driven e-commerce settings by highlighting the crucial role that customer attitudes play in shaping AI-related perceptions into buying intentions.

Acknowledgment
The author sincerely thanks all the respondents who participated in the survey. Besides, heartfelt gratitude is extended to all stakeholders for their valuable support and assistance, which made a significant contribution to the completion of this research.

view full abstract hide full abstract
    • Figure 1. Proposed research model
    • Figure 2. Measurement and structural links after removing AAI1 (SmartPLS 4.0)
    • Figure 3. Structural model results (PLS-SEM)
    • Table 1. Descriptive statistics results (N = 458)
    • Table 2. Reliability, convergent validity, and multicollinearity (SmartPLS 4.0)
    • Table 3. Discriminant validity (HTMT ratios)
    • Table 4. Coefficient of determination (R²) and adjusted R²
    • Table 5. Effect sizes (f²) for key paths
    • Table 6. Direct effects and hypothesis tests (H1-H8)
    • Table 7. Specific indirect effects (H9-H15)
    • Table 8. Total/overall effects on AAI and PI
    • Conceptualization
      Le Thi Kim Hoa
    • Data curation
      Le Thi Kim Hoa
    • Formal Analysis
      Le Thi Kim Hoa
    • Methodology
      Le Thi Kim Hoa
    • Visualization
      Le Thi Kim Hoa
    • Writing – original draft
      Le Thi Kim Hoa
    • Writing – review & editing
      Le Thi Kim Hoa