LAUSR

Abstract Despite progress in energy efficiency technologies and energy intensity reduction, the buildings sector's final energy consumption and associated CO2 emissions both continue to grow, pointing to the necessity of understanding real occupant behavior. This study, therefore, explores occupants' energy-related behavior in residential buildings with empirical data from a large-scale survey in Beijing, China, aiming to present an in-depth and comprehensive picture of occupant behavior. We obtained a total of 1003 valid responses to an online questionnaire concentrating on occupant behaviors with respect to space heating and cooling, water heating, cooking, lighting, appliances and other equipment. Our results show that, typically, purchase behavior is overall energy-efficient, as most appliances purchased are energy efficient. For both purchase and usage behaviors, air conditioning (AC), which is used for both space heating and cooling, plays a critical role as residents usually have less energy efficient AC in their homes and use it more often and for a longer period of time than other devices. Their habitual reactions are consistent across all habitual categories, but show no significant correlation with their purchase behavior. There is no single, straightforward and coherent pattern that can be explained by any single socio-demographic factor. These findings offer a behavioral insight into efforts to improve residential energy efficiency at the city scale. For instance, more policy attention should be directed to occupants’ purchase and usage behavior of ACs, such as by promoting the energy rating program in a more effective way. Moreover, given the lack of apparent linkage between purchase and habitual behaviors, a comprehensive policy framework incorporating different measures for different types of behavior is necessary. The complexity underlying behavioral choices also suggests the necessity of better understanding the behavioral patterns and their determining factors, as well as avoiding any simplistic assumption in policy making that aims to take behavioral factors into account.