Data for At the Bio-Digital Crossroads: Governance as a Conditioning Factor in AI-Driven Bioeconomy Systems
Description
This dataset contains the balanced panel data (2010–2021) utilized for the empirical analysis in Chapter 4, "At the Bio-Digital Crossroads: Governance as a Conditioning Factor in AI-Driven Bioeconomy Systems," focusing on four biodiversity-rich Latin American countries: Bolivia, Ecuador, Honduras, and Mexico. The underlying data is structured into four core analytical blocks: Socioeconomic (Real GDP per capita, household consumption expenditure), Environmental pressure (CO2 emissions/ecological footprint, water withdrawal per capita), Biocultural systems (indigenous population, biocultural savings), and Services/Welfare (life expectancy at birth, political and social participation). It includes the raw indicators, the Min-Max scaling scripts, and the Principal Component Analysis (PCA) outputs used to construct the governance composite indices (Techno-Governance [TG], Democratic Governance [DG], and Technological Governance [TechG]). These indices serve as the conditioning factors to analyze the bio-digital intersection, digital extractivism dynamics, and sustainability outcomes through fixed effects modeling.This dataset contains the balanced panel data (2010–2021) utilized for the empirical analysis in Chapter 4, "At the Bio-Digital Crossroads: Governance as a Conditioning Factor in AI-Driven Bioeconomy Systems," focusing on four biodiversity-rich Latin American countries: Bolivia, Ecuador, Honduras, and Mexico.
Files
Steps to reproduce
To reproduce the balanced panel dataset (2010–2021) for Bolivia, Ecuador, Honduras, and Mexico, follow the exact methodological steps described in Chapter 4 and Appendix A: 1. Data Collection and Sources: Gather the raw indicators for the four analytical blocks from their respective primary sources: - Socioeconomic Block: Real per capita GDP (constant 2015 US$) and Household consumption expenditure from the World Bank. - Environmental Block: CO2 emissions/ecological footprint from FAOSTAT and Global Forest Watch; Water withdrawal per capita from FAOSTAT. - Biocultural Block: Biocultural savings and Indigenous population from National Censuses and World Bank data. - Services / Welfare Block: Life expectancy at birth from the World Bank/UNDP; Political and social participation from V-Dem/World Bank. 2. Data Treatment and Normalization: - To account for missing values across the 2010–2021 time series, apply linear interpolation and backward/forward extrapolation techniques to ensure a balanced panel. - Apply a logarithmic transformation (Log-transformed) to variables requiring scaling stabilization, specifically Real per capita GDP and CO2 emissions. - Standardize all remaining indicators with disparate units using the Min–Max scaling method to map values strictly into a [0, 1] range. 3. Institutional Matrix and Composite Index Construction: - Construct the multi-dimensional Institutional Matrix using the normalized indicators. - Execute a Principal Component Analysis (PCA) on this matrix to reduce dimensionality and calculate orthogonal eigenvector weights. - Group the linear combinations to construct the three core composite indices: Techno-Governance (TG), Democratic Governance (DG), and Technological Governance (TechG). 4. Econometric Modelling Specification: - Structure the final data into a balanced panel model. - Execute a Fixed Effects (FE) specification to control for unobserved country-specific time-invariant heterogeneity across Bolivia, Ecuador, Honduras, and Mexico. - Set the composite sustainability outcomes derived from the PCA as the dependent variable, the three governance indices (TG, DG, TechG) as the independent variables, and Log-transformed GDP per capita as the macro-economic control variable.
Institutions
- Centro Agronomico Tropical de Investigacion y Ensenanza CatieCartago Province, Turrialba
- Universidad Mayor de San AndrésLa Paz Department, La Paz