Source data and reproducibility materials for Support definitions alter vulnerability alignment and country screening in adaptation finance

This repository provides the source data, analytical code, figures and reproducibility materials for the manuscript “Support definitions alter vulnerability alignment and country screening in adaptation finance”. The study examines whether conclusions about the alignment of bilateral adaptation support with climate vulnerability depend on how adaptation activities and financial values are defined. The analysis uses a harmonised panel of 128 non high income recipient countries and combines OECD Creditor Reporting System records for 2022 to 2024 with the 2023 ND GAIN vulnerability index, World Bank population data and World Bank country classifications. Ten support operationalisations are evaluated. These include one broad sectoral comparator and nine measures based on OECD adaptation markers, with support valued using nominal disbursements, grant only disbursements and grant equivalent values. All measures are constructed from the same universe of bilateral ODA grants and ODA loans with positive gross disbursements. The analytical workflow includes Spearman rank correlations, paired country bootstrap contrasts, multivariable regression with HC3 robust standard errors, false discovery rate adjustment, median and residual country screening, percentile threshold sensitivity, Jaccard similarity, country switching and selection frequency analysis across 100 specifications. The results show that marker based measures have stronger rank based alignment with vulnerability than the broad sectoral comparator. Grants and grant equivalent valuation also alter rank based alignment relative to nominal valuation, while direct differences between adjusted vulnerability coefficients remain uncertain. Country screening contains a small robust core, but the associated population estimate is highly sensitive to the classification rule. Angola, Eritrea, Myanmar and Sudan are selected in all 100 specifications. The repository includes processed country level data, source data for the main and supplementary figures, analysis scripts, figure reconstruction code, metadata, a data dictionary, citation information, quality assurance records and SHA 256 manifests. Grant equivalent values are interpreted as an OECD measure of donor effort. They are not treated as measures of recipient benefit, project performance or adaptation effectiveness. The analysis covers bilateral ODA only and does not estimate total adaptation finance, financing adequacy, donor intent or causal effects on vulnerability. Repository version: R8.1.1