data for PONE-D-25-41664

README for Dataset: "Unveiling Nonlinear Effects of Digital Inclusive Finance on Urban–Rural Integration A Threshold Panel Analysis of China"1. General DescriptionThis dataset contains provincial-level panel data for mainland China from 2011 to 2023, designed to investigate the impact of Digital Inclusive Finance (DIF) on Urban-Rural Integration (URI). The dependent variable is a composite Urban-Rural Integration (URI) index, constructed from 31 indicators across five dimensions: economic linkage, social equalization, spatial connectivity, ecological coordination, and demographic restructuring. The core explanatory variable is the Digital Inclusive Finance Index (DIF) developed by the Digital Finance Research Center at Peking University.<br>The dataset supports analyses of nonlinear effects (via threshold regressions), regional heterogeneity (eastern, central, western China), and robustness checks using alternative measures of digital economy development, traditional finance, and urban-rural decomposition of control variables.<br>2. File StructureFilename: Unveiling Nonlinear Effects of Digital Inclusive Finance on Urban–Rural Integration A Threshold Panel Analysis of China-data.xlsxFormat: Comma-separated values (xlsx)Rows: 403 observations (31 provinces × 13 years, with some missing due to data availability)Columns: 23 variables (see Section 3)3. Variable DefinitionsVariable Name Descriptionregion Geographic region: 1 = Eastern, 2 = Central, 3 = Westernprovince Province name (e.g., Beijing, Guangdong)year Calendar year (2011–2023)DIF Digital Inclusive Finance Index (overall score, standardized)D1 Coverage breadth dimension of DIFD2 Usage depth dimension of DIFD3 Digitalization level dimension of DIFGOV Fiscal support for agriculture: agricultural fiscal expenditure / agricultural outputOPEN Trade openness: total trade (exports + imports) / GDPEDU Education development: education expenditure / GDPFCI Fixed capital investment: total fixed asset investment / GDPMPP Mobile phone penetration: mobile subscriptions per 100 peopleURI Dependent variable: Composite Urban-Rural Integration Index (weighted sum of 31 standardized indicators)TF Traditional Financial Development Index (PCA-based on bank branches and employees)DE Digital Economy index (alternative robustness measure)Financial_Index_PCA Principal component used to construct TF (included for transparency)GDP_percapita Per capita GDP (log-transformed in analysis; raw value provided)Internet_penetration Internet users per 100 people (used to split high/low ICT groups)Tertiary_share Share of tertiary (service) sector in GDP (%)Agri_fiscal_share Share of agricultural fiscal expenditure in total fiscal expenditure (%)urban_rate Urbanization rate (% of population living in urban areas)edu_total Total education expenditure (in RMB, nominal)exports Total exports (in USD or RMB, nominal)export_share Export intensity: exports / GDPNote: All ratio variables (e.g., GOV, OPEN, EDU) are expressed as proportions (not percentages), unless otherwise noted.<br>4. Abbreviations and AcronymsURI: Urban-Rural IntegrationDIF: Digital Inclusive FinanceD1/D2/D3: Sub-dimensions of DIF (Coverage, Usage, Digitalization)GOV: Government fiscal support for agricultureOPEN: Trade opennessEDU: Education expenditure intensityFCI: Fixed Capital Investment intensityMPP: Mobile Phone PenetrationTF: Traditional Financial Development IndexDE: Digital Economy indexPCA: Principal Component AnalysisICT: Information and Communications Technology5. Data SourcesDIF Index (D1, D2, D3, DIF): Digital Finance Research Center, Peking UniversityMacroeconomic &amp; Fiscal Data: China Statistical Yearbook, Provincial Statistical YearbooksBanking Data (for TF): China Financial Statistics YearbookEducation &amp; Population Data: Ministry of Education, National Bureau of StatisticsTrade &amp; Exports: China Customs, WTO databasesInternet &amp; Mobile Data: China Internet Network Information Center (CNNIC), MIITAll raw data were processed by the authors. Missing values (e.g., Tibet in early years) are coded as NA.<br>6. Methodological NotesURI Construction: 31 indicators standardized (z-scores), then aggregated via entropy-weighting or expert-weighting (method documented in associated paper). Final URI is scaled to [0,1] or mean=0, sd=1 depending on version—this dataset uses standardized scores.TF Construction: Natural log transformation applied to bank branches and employees, followed by PCA. The first principal component (Financial_Index_PCA) is saved; TF is this component rescaled for interpretability.Interaction Terms: Not pre-computed in this file. Users should generate:edu_urban = edu_total * urban_rateedu_rural = edu_total * (1 - urban_rate)open_urban = OPEN * urban_rateRegional Classification: Based on China’s official regional grouping (Eastern: 11 provinces; Central: 8; Western: 12).7. Software RequirementsThis dataset is provided in .xlsx format and can be opened with:<br>Statistical software: R, Stata, Python (pandas)Spreadsheet programs: Microsoft Excel, Google Sheets, LibreOffice Calc<br>8. Usage RecommendationsUse province and year as panel identifiers.Handle missing values (NA) appropriately—especially for western provinces in early years.For threshold regression, use D1, D2, or D3 as threshold variables.To replicate robustness checks:Replace DIF with DE for digital economy robustness.Split sample by region, Internet_penetration (median split), GDP_percapita (terciles), or Tertiary_share.Interact Agri_fiscal_share with DIF to test public expenditure synergy.9. CitationIf you use this dataset, please cite the associated publication (to be added upon acceptance) and acknowledge the Digital Finance Research Center at Peking University for the DIF index.<br>10. ContactFor questions about the data or methodology, please contact:<br>Ying SongGreen Development Research Center, Wuyi University, Fujian, [e-mail] updated: November 2025