Sun Ang, Xiang Wang, Jiang Xu
School of Finance, Renmin University of China, Beijing, China.
Sci Rep. 2024 Mar 14;14(1):6169. doi: 10.1038/s41598-024-56852-2.
Extensive research has focused on the impact of weather on working capacity and income. However, in regions where income data largely relies on surveys, a pivotal yet underexplored question is whether weather not only influence real income but also introduce biases into survey-collected income data. We analyze longitudinal data from the China Health and Nutrition Survey and corresponding weather records from the Global Surface Summary of the Day, and uncover a negative correlation between survey-day temperature and self-reported annual income from the previous year. With a series of robustness checks, we confirm that the effect is primarily driven by behavioral factors rather than actual income changes. And threshold regression analyses show that the impact of temperature is more pronounced on hot days and relatively subdued or even reversed on cooler days. Further analyses indicate that mood, rather than cognitive capacity, plays a central role in causing the observed downward bias.
大量研究聚焦于天气对工作能力和收入的影响。然而，在收入数据很大程度上依赖调查的地区，一个关键但未被充分探索的问题是，天气是否不仅影响实际收入，还会给调查收集的收入数据带来偏差。我们分析了中国健康与营养调查的纵向数据以及来自全球日地面摘要的相应天气记录，发现调查当天的温度与前一年自我报告的年收入之间存在负相关。通过一系列稳健性检验，我们证实这种影响主要是由行为因素驱动的，而非实际收入变化。阈值回归分析表明，温度的影响在炎热天气更为显著，在较凉爽天气则相对较弱甚至相反。进一步分析表明，情绪而非认知能力在导致观察到的向下偏差中起核心作用。
