SPATIOTEMPORAL STATISTICAL ANALYSIS OF STUDENTS' SEAT LOCATION AND PSYCHOLOGICAL CONDITION DURING A LECTURE

Realizing a healthy classroom environment is an essential task in sustainable development. In particular, students acquire professional skills at universities and proper design of the classroom environment is important for improving the academic performance of university students. However, there are many cases where the classroom environment is not ideal, such as when doors or windows are not opened. Many studies have analyzed the relationships between classroom environment and students' academic performance. However, few have analyzed the relationships between the spatiotemporal distribution of the classroom thermal environment and temporal changes in students' psychological sensations or concentration levels. Therefore, the main theme of this study is the exploration of the relationships between these factors. The survey was conducted in a university classroom in May 2013. All doors and windows in the classroom were closed to avoid the effects of ventilation and create the hot, humid condition of an actual classroom. The thermal environment was continuously measured at several points during the 90-min lectures, and students were asked to complete a questionnaire on their psychological sensations and concentration levels every 15 min; this allowed for collection of both physical-psychological data and spatiotemporal data of the classroom thermal environment. In the previous paper in this study (Part 1), the problems of a spatiotemporally heterogeneous thermal environment in a large classroom and students' psychological condition during a lecture were clarified. In particular, the back of the classroom tended to be hot. In this paper (Part 2), psychological questionnaire data were reanalyzed in detail from the perspective of spatiotemporal statistical analysis, which are frequently applied to city-scale studies. Local spatiotemporal autocorrelation statistics (Getis-Ord Gi*) revealed significantly hot and cold spots. The results of the analysis were visualized on 3D spatiotemporal maps. The results revealed that the number of cold spots for students' thermal sensation or tiredness level and the number of hot spots for students' concentration level in some seats in the front of the classroom decreased toward the end of the lecture. In contrast, the number of hot spots for students' thermal sensation or tiredness level and the number of cold spots for students' concentration level in some seats in the back of the classroom increased with time during the lect