Heat and Helping

<b>Update Note (16 Dec 2024):</b>During the peer review of a paper reporting these results, data files "data_alt.csv" and "data_alt_r1.csv" were hosted in this domain for reviewers to evaluate the work. Since the data contain copyrighted information, they cannot be shared publicly. Therefore, upon acceptance of publication of the work (15 Dec 2024), the two datasets were removed from this domain. Colleagues who are interested in the data can contact the corresponding author (Henry Ng: [e-mail]) for more information. For security reasons, ipynb files containing the author's directory paths have been removed as well.<b>Data Files (removed):</b>data_alt.csv is the data reported in the first submission of this project. data_alt_r1.csv is an expanded data file in response to comments after the first review.data_overview summarizes the variable columns of data files data_alt.csv and data_alt_r1.csv.<b>Results Files:</b>ms_results.ipynb hosts summary statistics and visuals reported in the results section of the published article.donate_m1_m6.ipynb, help_strngr_m1_m6.ipynb, and volunteer_m1_m6.ipynb host the PyStan codes that generate the MCMC chains for Bayesian inferences of Models 1 to 6 for each of the three outcome variables.<b>Supplementary Files:</b>supplementary_r1_1 corresponds to additional analyses required by R1C3, R2C5 and R2C4, regarding the use of the metric form of GNIPC instead of the dummy coded version. It also explores three covariates, namely GINI index, ghg emission, and % of population attaining tertiary education.supplementary_r1_2 corresponds to additional analyses required by R3C3, regarding the adoption of different priors. Four additional priors have been tested to show the minimal impact of the choice of priors on the analysis.supplementary_ar1_(hlp/don/vol) corresponds to analyses of Model 3, 5, and 6, with the addition of an autoregressive process (AR1) in the time series. The analyses report findings consistent with the original analyses without the AR1 process.

更新说明（2024年12月16日）：在一篇报道本研究成果的论文同行评审期间，本域名曾托管data_alt.csv与data_alt_r1.csv两份数据文件，供评审专家评估该项研究工作。因上述数据包含受版权保护的内容，无法对外公开共享。因此，在该论文于2024年12月15日被录用发表后，这两份数据集已从本域名中移除。有意获取该数据的同行可联系通讯作者（Henry Ng: [e-mail]）以获取更多相关信息。出于安全考量，包含作者目录路径的ipynb文件也已一并移除。
已移除的数据文件：data_alt.csv为本项目首次投稿时所使用的数据集；data_alt_r1.csv为响应首次评审意见后扩充得到的数据集；data_overview汇总了data_alt.csv与data_alt_r1.csv的数据列变量信息。
结果文件：ms_results.ipynb包含已发表论文结果部分中报告的描述性统计量与可视化图表；donate_m1_m6.ipynb、help_strngr_m1_m6.ipynb以及volunteer_m1_m6.ipynb包含PyStan代码，用于生成针对三个因变量分别构建的1至6号模型的贝叶斯推断马尔可夫链蒙特卡洛（MCMC）链。
补充文件：supplementary_r1_1对应评审意见R1C3、R2C5与R2C4要求开展的补充分析，该分析采用GNIPC的度量形式而非虚拟编码形式，并额外考察了三项协变量：基尼系数（GINI index）、温室气体排放（ghg emission）以及高等教育普及率。supplementary_r1_2对应评审意见R3C3要求开展的补充分析，该分析采用了不同先验分布。本次共测试了四种额外先验，以验证先验选择对分析结果的影响极小。supplementary_ar1_(hlp/don/vol)对应在时序模型中加入自回归过程（AR1）后针对3号、5号及6号模型开展的分析，结果显示其与未加入AR1过程的原始分析结论保持一致。