The Effect of Structural Equation Modeling on Chatbot Usage: An Investigation of Dialogflow

TL;DR

This study investigates how users perceive the Dialogflow chatbot framework, examining factors like service awareness and output quality, and their influence on user acceptance using Structural Equation Modeling.

Contribution

It applies Generalized Structured Component Analysis to validate the Technology Acceptance Model in the context of Dialogflow, providing new insights into user perceptions.

Findings

Perceived ease of use and usefulness significantly influence behavioral intention.

Service awareness and output quality predict perceived usefulness.

Task-technology fit positively impacts perceived ease of use.

Abstract

This study aims to understand users' perceptions of using the Dialogflow framework and verify the relationships among service awareness, task-technology fit, output quality, and TAM variables. Generalized Structured Component Analysis was employed to experiment with six hypotheses. Two hundred twenty-seven participants were recruited through the purposive non-random sampling technique. Google Forms was utilized as a medium to develop and distribute survey questionnaires to subjects of interest. The experimental results indicated that perceived ease of use and usefulness had a statistically significant and positive influence on behavioral intention. Awareness of service and output quality was considered reliable predictors of perceived usefulness. Also, perceived task-technology fit positively affected perceived ease of use. The model specification accounted for 50.04% of the total variation. The findings can be leveraged to reinforce TAM in future research in a comparative academic context to validate the hypothesis. Several practitioner recommendations and the study's limitations have been presented.