Dataset and code: "Revisiting the Social Trust–Economic Growth Nexus: A Nonlinear Approach"

================================================================================DATASET AND CODE REPOSITORY"Revisiting the Social Trust–Economic Growth Nexus: A Nonlinear Approach"Humanities and Social Sciences Communications (Second Revision, June 2026)================================================================================ CONTENTS--------1. merged_trust_growth_dataset_FINAL.xlsx — Panel dataset2. trust_growth_analysis.R — Full analysis code (R)3. README.txt — This file --------------------------------------------------------------------------------1. DATASET: merged_trust_growth_dataset_FINAL.xlsx-------------------------------------------------------------------------------- Sheet: Merged_Data The main analytical dataset. Each row is one country–WVS wave observation. 98 countries, WVS Waves 3–7 (1995–2022), 357 rows before listwise deletion, 266 complete observations used in the main analysis. Columns: Entity : Country name Code : ISO 3166-1 alpha-3 country code Year : WVS wave year (1998, 2004, 2010, 2014, 2022) logGDPGrw : log(1 + period-avg annual GDP per capita growth / 100) Source: WDI NY.GDP.PCAP.KD.ZG GDPpcGrowth : Period-average annual GDP per capita growth rate (%) Source: WDI NY.GDP.PCAP.KD.ZG (pre-log version) Trust : Share of WVS respondents: "most people can be trusted" Source: World Values Survey Waves 3–7 GFCF : Gross fixed capital formation (% of GDP), period avg Source: WDI NE.GDI.FTOT.ZS GCF : Gross capital formation (% of GDP), period avg Source: WDI NE.GDI.TOTL.ZS FDI : FDI net inflows (% of GDP), period avg Source: WDI BX.KLT.DINV.WD.GD.ZS PopGrowth : Population growth (annual %), period avg Source: WDI SP.POP.GROW HumanCapital : Gross tertiary school enrollment (%), period avg Source: WDI SE.TER.ENRR / UNESCO Institute for Statistics Corruption : Control of Corruption percentile (0–100) Source: World Governance Indicators CC.EST ExternalBalance : External balance on goods and services (% of GDP), period avg Source: WDI NE.RSB.GNFS.ZS Period-averaging rule: WDI and WGI variables are annual; WVS data are wave-based. For each country–wave, WDI values are averaged over the corresponding window: Wave 3 (Year = 1998): average of annual values 1995–1998 Wave 4 (Year = 2004): average of annual values 1999–2004 Wave 5 (Year = 2010): average of annual values 2005–2010 Wave 6 (Year = 2014): average of annual values 2011–2014 Wave 7 (Year = 2022): average of annual values 2017–2022 Note on South Korea (KOR): External balance data were absent from the original WDI batch download. Values were supplemented directly from WDI series NE.RSB.GNFS.ZS and integrated using the same period-averaging procedure. Cells filled in this way are highlighted in green in the Merged_Data sheet. Sheet: Missing_Check Country-level summary of missing observations for each analysis variable. Sheet: Variable_Definitions Full variable definitions, WDI series codes, sources, and roles in the model. Sheet: Period_Mapping Explicit mapping of WVS wave years to WDI averaging windows. --------------------------------------------------------------------------------2. ANALYSIS CODE: trust_growth_analysis.R-------------------------------------------------------------------------------- Language : R (version 4.x or later)Required packages: readxl, dplyr, tidyr, plm, pgmm, lmtest, sandwich, car, ggplot2, ggcorrplot, stargazer, AER To install all packages at once: install.packages(c("readxl","dplyr","tidyr","plm","pgmm","lmtest", "sandwich","car","ggplot2","ggcorrplot","stargazer","AER")) To run: 1. Place trust_growth_analysis.R and merged_trust_growth_dataset_FINAL.xlsx in the same directory. 2. Set working directory to that folder: setwd("path/to/folder") 3. Run: source("trust_growth_analysis.R") The script produces, in order: - Descriptive statistics (Table 2) - Correlation matrix with significance levels (Table 4) - VIF diagnostics (Tables 5) - Panel IV (2SLS) results (Table 3) - Panel Threshold Regression — single and double threshold (Table 6) - Quadratic LSDV model (Table 7) - Multiple threshold models (Table 8) - Robustness checks R1–R4b (Table 9) - Country composition by trust regime - Stargazer summary table Estimated run time: 3–5 minutes (bootstrap with 500 replications). --------------------------------------------------------------------------------3. DATA SOURCES-------------------------------------------------------------------------------- World Values Survey (WVS) Waves 3–7, 1995–2022 Haerpfer, C., Inglehart, R., Moreno, A., Welzel, C., Kizilova, K., Diez-Medrano J., Lagos, M., Norris, P., Ponarin, E. & Puranen, B. (eds.) (2022). World Values Survey: Round Seven – Country-Pooled Datafile Version 5.0. Madrid, Spain & Vienna, Austria: JD Systems Institute & WVSA Secretariat. https://www.worldvaluessurvey.org World Bank World Development Indicators (WDI) World Bank (2025). World Development Indicators. https://databank.worldbank.org/source/world-development-indicators Series used: NY.GDP.PCAP.KD.ZG, NE.GDI.FTOT.ZS, NE.GDI.TOTL.ZS, BX.KLT.DINV.WD.GD.ZS, SP.POP.GROW, SE.TER.ENRR, NE.RSB.GNFS.ZS World Governance Indicators (WGI) World Bank (2025). Worldwide Governance Indicators. https://www.worldbank.org/en/publication/worldwide-governance-indicators Series used: CC.EST (Control of Corruption) --------------------------------------------------------------------------------4. CITATION-------------------------------------------------------------------------------- If you use this dataset or code, please cite the accompanying article: Okuyan, H. A. & Altan, İ. M. (2026). Revisiting the Social Trust–Economic Growth Nexus: A Nonlinear Approach. Humanities and Social Sciences Communications. doi: [to be assigned upon publication] Dataset and code: Okuyan, H. A. & Altan, İ. M. (2026). Dataset and code: "Revisiting the Social Trust–Economic Growth Nexus: A Nonlinear Approach" [Data set]. Zenodo. https://doi.org/10.5281/zenodo.15715548 --------------------------------------------------------------------------------LICENSE-------------------------------------------------------------------------------- Dataset: Creative Commons Attribution 4.0 International (CC BY 4.0)Code : MIT License The underlying survey and indicator data retain the licenses of theirrespective original sources (WVS, World Bank).================================================================================