Panel Dataset on Energy Transition, Digitalisation, Governance and Greenhouse Gas Emission Nexus in Sub-Saharan Africa (2000-2023)

Description

This dataset contains a country-year panel for Sub-Saharan Africa covering 2000-2023, assembled for research on the nexus between energy transition, digitalisation, governance, and greenhouse gas emissions. It includes indicators of energy intensity, electricity access, renewable energy use, digital connectivity, economic development, urbanisation, and GHG emissions in formats suitable for panel-data analysis, replication, and visualisation. The main analytical file, SSA_energy_digital_emissions_panel_data_2000_2023.csv, can be used to run the empirical analysis. The companion documentation file, SSA_energy_digital_governance_emissions_panel_codebook.xlsx, provides the variable codebook, dataset structure, and relevant processing notes, including missingness and related data-handling information. A key feature of the dataset is its alignment with the study’s empirical design. The panel file supports the baseline, moderation, heterogeneity, and forward-looking analyses, while the z-score file, SSA_energy_digital_emissions_zscore_data_mean _2000_2023.csv, is used to generate the country heat map of variable patterns. The separate Africa-wide file, GHG Mean by Country in Africa 2020_2023.csv, is used for the African GHG choropleth map. Together, these files support both the main econometric analysis and the study’s descriptive visualisations. Files included: SSA_energy_digital_emissions_panel_data_2000_2023.csv SSA_energy_digital_governance_emissions_panel_codebook.xlsx SSA_energy_digital_emissions_zscore_data_mean _2000_2023.csv GHG Mean by Country in Africa 2020_2023.csv Source: World Bank, World Development Indicators (WDI).

Steps to reproduce

Step 1 - Data acquisition: Download annual country-level data from the World Bank’s World Development Indicators (WDI) for Sub-Saharan Africa covering 2000–2023. The study uses the following indicators: ghg = total greenhouse gas emissions excluding LULUCF; ein = energy intensity level of primary energy; elec = access to electricity; ren = renewable energy consumption; mob = mobile cellular subscriptions; inte = individuals using the Internet; gov = government effectiveness-estimate; gdppc = GDP per capita; and urban = urban population as a share of total population. Step 2 - Country coverage: Restrict the sample to 46 Sub-Saharan African countries. Eritrea and South Sudan are excluded because extensive missing values prevent consistent panel construction and reliable interpolation across the study horizon. Step 3 - Data import and harmonisation: Import the downloaded series into Stata and merge them by country code and year. Retain only the study variables and rename them to the analysis labels: ghg, ein, elec, ren, mob, inte, gov, gdppc, urban. Then create a numeric panel identifier using egen cid = group(iso3c) and sort by cid year. Step 4 - Study design rule: Keep the panel for 2000–2023, but retain ghg through 2023 while restricting all non-GHG variables to 2021. In Stata: foreach v in ein elec ren mob inte gov gdppc urban { replace v' = . if year > 2021 }` Step 5 - Missing-data treatment: Within each country, fill missing observations using linear interpolation with: ipolate. Apply this to the raw series where needed, then re-impose the 2021 cutoff for all non-GHG variables so that only ghg remains available through 2023. Step 6 - Variable construction: Generate the transformed analytical variables used in the study: gen ln_ghg = ln(ghg) gen ln_ein = ln(ein) gen ln_gdppc = ln(gdppc) Construct the digitalisation index (dig) as a z-score composite of mob and inte. Then create country-mean centred variables for the moderation models, including centred versions of ln_ein, ren, elec, dig, and gov, and generate the required interaction terms. Step 7 - Replication-ready outputs: Save the processed analytical file as the final panel dataset. Prepare the companion codebook/documentation file reporting variable definitions, country codes, transformations, and missingness before and after interpolation. Step 8 - Visualisation files: From the final panel, generate the country mean z-score file used for the heat map and the separate Africa GHG country file used for the choropleth map.

Institutions

• Asia Pacific University of Technology & Innovation

  Kuala Lumpur, Kuala Lumpur
• Universiti Putra Malaysia

  Selangor, Seri Kembangan

Categories

Licence