Causal effects estimation: Using natural experiments in observational field studies in building science

Abstract

Correlational analysis, such as linear regression, does not imply causation. This paper introduces and applies a causal inference framework and a specific method, regression discontinuity, to thermal comfort field studies. The method utilizes policy thresholds in China, where the winter district heating policy is based on cities' geographical locations relative to the Huai River. The approximate latitude of the Huai River can be considered as a natural, geographical threshold, where cities near the threshold are quite similar, except for the availability of district heating in cities north of the threshold, creating a situation similar to a randomized experiment. Using the regression discontinuity method, we quantify the causal effects of the experiment treatment (district heating) on the physical indoor environments and subjective responses of building occupants. We found that mean indoor operative temperatures were 4.3 °C higher, and mean thermal sensation votes were 0.6 warmer due to the district heating. In contrast, using conventional correlational analysis, we demonstrate that the correlation between indoor operative temperature and thermal sensation votes does not accurately reflect the causal relationship between the two. We also show that the indoor operative temperature could be either positively or negatively correlated with occupants’ thermal satisfaction. However, we cannot conclude that increasing the indoor operative temperature in these circumstances will necessarily lead to higher or lower thermal satisfaction. This highlights the importance of causal inference methods in thermal comfort field studies and other observational studies in building science, where the regression discontinuity method might apply.