*3.1. Study Overview*

This research conducted a field experiment in Xi'an, one of the most densely populated cities located in Shaanxi Province, northwest China, and selected two typical residential communities in Yan-Ta district (urban area in Xi'an) as the study areas. Notably, these two communities are located in the same district and the residents in the selected study areas share similar socioeconomic characteristics. Especially, the participants have similar cultural and economic backgrounds as well as a built surrounding environment. Before substantial data gathering, the researchers invited five experts to review as well as revise the questionnaire in terms of its structure, wording and comprehensibility to make sure the questionnaire was understandable for the general public. These experts included two researchers in psychology, two experienced community workers in the study areas, as well as one property manager. The revised questionnaire was pilot-tested by 30 individuals to give revision suggestions to the final version of the questionnaire.

In this study, the community workers played the role as the gatekeeper. The study used the two-stage sampling method. Residents in the selected communities were encouraged to participate in this research by the staff in the neighborhood committee via a Wechat group to maximize the sample size. The researchers called for participants in the study areas through the local residential committee and there were 853 households signed-up. The researchers selected 800 households from the signed-up

group. The selection was based on two standards: (1) the participant had lived in the investigation sites for at least six months; and (2) the participant had a steady job and had worked at the current company for at least six months so that they are familiar enough with the environments.

The participants in the experiment were purposely divided into two groups (i.e., Group A and Group B) according to their sociodemographic backgrounds to maximize the representativeness of the diverse background and to reduce the possibility of sampling bias. The respondents in di fferent groups were required to complete the di fferent questionnaires with di fferent themes. The first version of the questionnaire concerns HESB (so called HESB-Q) and was prepared for Group A. The second version of the questionnaire concerns OESB (so called OESB-Q) and was prepared for Group B.

The responses from the participants were collected during January to March 2020. Due to the coronavirus (i.e., COVID-19) outbreak in China, the researchers adopted online tools to avoid face-to-face interaction from late January. Especially, the researchers employed Wechat, one of the most popular instant message Apps for consumer smart devices (i.e., smart phones and tablets) in China, to distribute the questionnaire and to collect the responses. The high usage rate makes Wechat (7.0.12, Tencent, Shenzhen, China) a suitable tool to approach potential participants in Chinese cities [8,43,51]. Sun et al. [51] provide sufficient evidence showing the feasibility of questionnaire distribution with Wechat. In this study, potential respondents were sent the online questionnaire which is able to be opened by clicking the link or scanning a QR-code on a prepared survey card to reach the online questionnaire. The structural equation modeling (SEM) approach is then adopted to delineate the causal mechanisms and numerous relationships between personality traits and di fferent types of energy-saving behaviors in this study.
