global cultural policy landscape

This dataset accompanies the article "Decoding the Global Landscape of Cultural Policy: A Data-Driven Analysis Using Unsupervised Learning" (Park & Chang). It covers 179 nations (ISO 3166-1) and combines three types of variables used to derive an empirical typology of national cultural policy regimes via principal component analysis and K-means clustering. First, socioeconomic indicators: GDP per capita, PPP (World Bank, 2023); urban population share (World Bank, 2023); and the Human Development Index (UNDP, 2023 values). Second, cultural policy duration variables (author-collected): the historical depth of six categories of cultural and creative industry (CCI) legislation, measured as years since original enactment (2025 minus the year of first introduction, tracing amended laws back to their earliest predecessor) — framework/basic culture laws, content-industry laws, CCI tax incentives, CCI entrepreneurship support, copyright protection, and artist welfare systems. A value of zero indicates no such instrument exists. Data were collected through exploratory search restricted to credible official sources (government portals and international-organization databases); non-English legislative texts were translated with LLM assistance and verified against official summaries. Entries were audited for false zeros and anachronistic enactment years, and provenance records (law names and source URLs) are included. Third, CCI market indicators used for post-hoc analysis only, due to missingness: creative services exports and creative goods exports as shares of total trade, and entertainment & media market size per 1,000 population aged 15–69, from WIPO's Global Innovation Index 2025 (indicators 7.2.1, 7.2.4, 7.2.3). The deposit also includes the final cluster assignments (k = 2, 3, 4) and a Python replication script reproducing the PCA, clustering, statistical tests, and robustness analyses reported in the paper. Funding: Ministry of Education of the Republic of Korea / National Research Foundation of Korea (NRF-2025S1A5C2A02022476).本数据集配套于论文《解码文化政策全球图景：基于无监督学习的数据驱动分析》（Decoding the Global Landscape of Cultural Policy: A Data-Driven Analysis Using Unsupervised Learning，Park与Chang合著）。本数据集覆盖179个遵循ISO 3166-1标准的国家，整合三类变量，通过主成分分析（Principal Component Analysis, PCA）与K-means聚类，推导得到国家文化政策体系的经验分类类型。 第一类为社会经济指标：按购买力平价（Purchasing Power Parity, PPP）计算的人均国内生产总值（Gross Domestic Product, GDP）（世界银行（World Bank），2023年数据）、城镇人口占比（世界银行（World Bank），2023年数据），以及人类发展指数（Human Development Index, HDI，联合国开发计划署（United Nations Development Programme, UNDP），2023年数据）。第二类为文化政策时长变量（由作者自行收集）：六类文化创意产业（Cultural and Creative Industry, CCI）立法的历史积淀程度，以首次颁布至今的时长测算（2025年减去首次立法年份，修订法案追溯至其最早的原始版本）——涵盖框架性/基础性文化法律、内容产业法规、文化创意产业税收优惠政策、文化创意产业创业扶持政策、版权保护制度以及艺术家福利体系。若取值为0，则代表未设立该类政策工具。数据通过定向检索可信官方来源（政府门户网站与国际组织数据库）收集完成；非英语立法文本借助大语言模型（Large Language Model, LLM）翻译，并对照官方摘要进行核验。研究人员对条目进行了审计，以排查虚假零值与不合时宜的立法年份，并附带了来源记录（法律名称与源URL）。第三类为文化创意产业市场指标：因存在数据缺失，仅用于事后分析，包括创意服务出口额与创意商品出口额占总贸易额的比重，以及每千名15-69岁人口对应的娱乐与媒体市场规模，数据源自世界知识产权组织（World Intellectual Property Organization, WIPO）《2025年全球创新指数》（指标编号7.2.1、7.2.4、7.2.3）。 本数据集还包含最终的聚类结果（k=2、3、4），以及复现论文中主成分分析、聚类、统计检验与稳健性分析的Python脚本。 资助信息：韩国教育部/韩国国家研究基金会（NRF-2025S1A5C2A02022476）。