The moderating effect of user-related characteristics between chatbot design factors and user trust in m-commerce chatbots

Date of Publication


Document Type

Bachelor's Thesis

Degree Name

Bachelor of Science in Industrial Engineering

Subject Categories

Engineering | Industrial Engineering


Gokongwei College of Engineering


Industrial Engineering

Thesis Advisor

Jazmin C. Tangsoc

Defense Panel Chair

Jose Edgar S. Mutuc

Defense Panel Member

Alma Maria Jennifer A. Gutierrez
Eric A. Siy


With the emergence of digitalization, chatbots are now being utilized to cater to consumers of diverse industries such as e-commerce, healthcare, and the like. Specifically, chatbots are being used as customer service to address queries in a quick and efficient manner. Nevertheless, since chatbots are relatively new in today’s society, user trust plays a vital role in translating to users’ intention to use this artificial intelligence-based technology. Previous studies have explored and emphasized the key factors that influence user trust in chatbots. These are namely anthropomorphism, responsiveness, expertise, risk, propensity to trust technology, and brand perception. However, past studies have yet to examine the moderating effect of propensity to trust technology and brand perception between chatbot design factors (anthropomorphism, responsiveness, expertise, and risk) and user trust extending to intention to use, and integrate various findings on the impacts of responsiveness and validate the significance of anthropomorphism on trust. To address these, the objectives of the study were as follows: (1) to determine the significance of anthropomorphism and responsiveness as chatbot design factors towards the user trust in chatbots, (2) to determine the moderating effect of propensity to trust technology and brand perception between chatbot design factors and user trust, (3) to identify the combination of chatbot design factors that will result in user trust, and (4) to determine whether user trust has a significant effect on users' intention to use m-commerce chatbots.

The factors utilized in this study were based on past research and were validated through one-on-one interviews. A design of experiment was conducted involving the 4 chatbot design factors with 2 levels forming a total of 16 design combinations. The experimentation phase includes the following: a pre-experimentation questionnaire to capture relevant participant data and controls, a customer service chatbot interaction with randomized design factor settings, and a post-experiment questionnaire to measure the dependent variables (user trust and intention to use). In obtaining the results, structural equation modeling (SEM) was utilized given that it is capable of identifying causal relationships between multiple variables in a simultaneous manner.

Results showed that anthropomorphism and expertise had a significant effect on users’ trust in chatbots. Specifically, the combination that yielded the highest levels of trust was low anthropomorphism (identifies as an automated customer service chatbot, possesses a human avatar, with conversational cues) and high expertise. Participants prefer a chatbot that is transparent about their identity and concrete in settling customer queries. It was also found that an individual’s propensity to trust technology has a moderating effect and strengthened the negative relationship between anthropomorphism and user trust. This indicates that users with a high trust propensity seek chatbots that introduce themselves as automated customer service chatbots because it shows honesty and transparency. On the other hand, brand perception did not have a moderating effect between design factors and user trust. For some participants, they viewed the chatbot and brand as separate entities, while others prioritize the performance of the chatbot over the brand. Lastly, the results of the study displayed that user trust significantly and positively affects users’ intention to use m-commerce chatbots. Thus, the creation of user-trusted chatbots is a promising venture as it is able to translate to the adoption of the technology.

Abstract Format







Human-robot interaction; Robotics—Human factors; Online chat groups; Consumers—Attitudes; Mobile commerce

Upload Full Text


Embargo Period


This document is currently not available here.