Causal Analysis for Thermal Comfort Votes

To analyze the causal relationship between comfort environment vote and thermal sensation votes (of face, chest, hands, buttocks, and soles of the foot), this paper presents how three methods work in experimentation: multiple regression analysis, conjoint analysis and structural equation modeling (SEM). The samples for the experimentation consisted of 1,584 (22 subjects x 9 conditions x 8 times) questionnaire data. Our main interest is whether there are any differences in the causal relationship among subjects, conditions or methods. Paying attention to preprocessing, we found the following: (1) the importance of each thermal sensation vote differed not only according to subject but also according to the analysis method; (2) SEM is the most suitable method to explain the causal relationship; and (3) the lower body is more important than the upper body in controlling thermal comfort.