electricsheepafrica/africa-ilo-ees-stem-sex-jbc-nb-employees-in-stem-occupations-by-sex-and-type-of-j
收藏资源简介:
该数据集名为“按性别和合同类型划分的STEM职业员工(千)| 非洲(ILOSTAT)”,是一个关于非洲地区STEM(科学、技术、工程和数学)职业就业情况的表格数据集。它包含1,053个观测值,覆盖34个非洲国家(如埃及、卢旺达、赞比亚等),时间跨度为2009年至2025年,每年提供数据。数据集核心指标为“EES_STEM_SEX_JBC_NB”,表示按性别和合同类型划分的STEM职业员工数量(以千为单位)。数据来源于国际劳工组织(ILO)的ILOSTAT数据库,通过REST API直接获取,并过滤到非洲国家代码。ILOSTAT使用国际劳工统计学家会议(ICLS)定义对原始调查微观数据进行标准化处理。数据集包含多列,如国家代码(ref_area)、国家名称(ref_area.label)、数据源(source)、指标代码(indicator)、性别分类(sex,包括总计、男性和女性)、合同类型分类(classif1)、年份(time)、观测值(obs_value)以及数据质量标志(如观测状态obs_status)。数据按年度频率发布,当同一国家×年份有多个数据源时,采用ILO选择的“最佳来源”。该数据集适用于机器学习任务,如表格分类、回归和时间序列预测,可用于分析非洲STEM劳动力市场的性别和合同结构趋势。数据集由Electric Sheep Africa重新打包,以Parquet格式发布,便于通过Hugging Face的datasets库加载使用。
This dataset is titled Employees in STEM occupations by sex and type of job contract (thousands) | Africa (ILOSTAT) and is a tabular dataset focusing on STEM (Science, Technology, Engineering, and Mathematics) occupational employment in Africa. It contains 1,053 observations across 34 African countries (e.g., Egypt, Rwanda, Zambia) spanning the years 2009 to 2025, with annual data. The core indicator is EES_STEM_SEX_JBC_NB, which represents the number of employees in STEM occupations disaggregated by sex and type of job contract (in thousands). Data is sourced from the International Labour Organization (ILO) ILOSTAT database, pulled directly via the REST API and filtered to African ISO3 country codes. ILOSTAT harmonizes raw survey microdata using International Conference of Labour Statisticians (ICLS) definitions. The dataset includes columns such as country code (ref_area), country name (ref_area.label), data source (source), indicator code (indicator), sex disaggregation (sex, including total, male, and female), contract type classification (classif1), year (time), observed value (obs_value), and data quality flags (e.g., observation status obs_status). Data is published at annual frequency, and when multiple sources exist for the same country×year, the ILO-selected best source is used. It is suitable for machine learning tasks like tabular classification, regression, and time-series forecasting, enabling analysis of trends in STEM labor markets across Africa by gender and contract type. The dataset is repackaged by Electric Sheep Africa in Parquet format for easy loading via the Hugging Face datasets library.




