MuniciMex-2020: sociodemographic data of municipalities of Mexico from the 2020 census

MuniciMex-2020 is produced from open data of the 2020 Population and Housing Census of Mexico (Mexican Institute for Statistics and Geography, INEGI, 2021). The aggregation unit is municipality, so the whole 2,469 municipalities are available in the dataset. In addition to the original 222 attributes in the official census database, percentages are appended. These percentages are computed from diverse counts of inhabitants and houses in the census. For each attribute that is a count of inhabitants or a count of houses regarding any socio-demographic aspect, a new attibute is created as a percentage. Percentages computed for inhabitants characteristics are calculated from the total municipal count of inhabitants. In turn, percentages computed for houses characteristics are calculated from the total municipal count of houses. For instance, the dataset contains percentage of inhabitants who know to read and write, percentage of houses that have sewage facilities, percentage of houses that have Internet service, etc. This percentages are highly useful to produce a broad series of statistical analyses and machine learning models because they constitute a statistical normalization enabling to perform comparisons among municipalities that have (very) different counts of inhabitants (or houses) among each other.MuniciMex-2020数据集源自墨西哥2020年人口与住房普查的公开数据（墨西哥国家统计与地理研究所，Mexican Institute for Statistics and Geography，缩写INEGI，2021）。本数据集的统计聚合单元为墨西哥市政区，完整涵盖全部2469个市政区的相关数据。除官方普查数据库中原有的222项属性外，数据集还附加了各类百分比衍生指标。此类百分比由普查中各类人口与住房计数数据计算得到：针对涉及社会人口学维度的各类人口计数或住房计数属性，均新增对应的百分比属性。其中，人口特征相关百分比以对应市政区的总人口数为计算基数，住房特征相关百分比则以对应市政区的总住房数为计算基数。举例而言，数据集包含具备读写能力的人口占比、拥有污水处理设施的住房占比、配备互联网服务的住房占比等指标。此类百分比可实现统计标准化，能够在人口（或住房）总量差异悬殊的各市政区之间开展横向对比，因此广泛适用于各类统计分析与机器学习模型构建工作。