Moderation Analysis of Gender in Social Robot Interactions

Abstract

The current study aimed to further analyze Schadenberg et al.’s (2021) dataset examining perceived social attributions after observing human-robot interactions. In their study, visibility of an external cause was found to significantly predict perceived competence of the Robot Social Attributes Scale (RoSAS) only. However, no demographic variable was considered in the analyzes. Therefore, gender was analyzed as a moderator between visibility and participants’ perceived competence of the robot. Additionally, a Confirmatory Factor Analysis was conducted on the RoSAS measure to examine its validity in the current context. Results indicated that gender did not significantly moderate the relationship between visibility and perceived competence. This finding could be explained by the RoSAS demonstrating poor fit with the data even though the measure indicated high internal reliability. In light of these results, the current study advocates for further psychometric validation of newer scales across varying conditions, especially within the social robotics field.

Publication
In Proceedings of the 66th Human Factors and Ergonomics Society International Annual Meeting.
Liam Kettle
Liam Kettle
UX & Human Factors Researcher