The Effects of empathy and personalization on chatbot usage intention : A structural equation modeling analysis based on the elaboration likelihood model (ELM)
This study investigates how social-affective features of chatbots, namely empathy and personalization, influence users’ intention to use them. Grounded in the elaboration likelihood model (ELM), empathy is conceptualized as a peripheral cue, while personalization functions as a central cue. Trust and perceived usefulness are proposed as mediators of these effects. Partial least squares structural equation modeling (PLSSEM) was applied to data collected from 289 respondents with chatbot experience. The results show that both empathy and personalization significantly impact trust and perceived usefulness but do not directly influence usage intention. Rather, their effects are fully mediated by trust and usefulness. Furthermore, a multi-group analysis reveals that users' prior chatbot experience moderates the relationships. The findings provide theoretical insights into affective persuasion mechanisms and practical implications for experience-based chatbot design.
목차
Abstract 1. Introduction 2. Theoretical background and hypotheses 2.1 Chatbot communication and persuasion framework 2.2 Key constructs and route classification 2.3 Hypotheses development 3. Research model 3.1 Structure of the model 3.2 Definition of variables 3.3 Analytical approach 4. Results of empirical analysis 4.1 Demographic profile of the sample 4.2 Evaluation of the measurement model 4.3 Structural model results 4.4 Mediation analysis 4.5 Multi-group analysis by usage experience 5. Conclusion and implications 5.1 Theoretical and practical implications 5.2 Limitations and Future Research References