Humanitarian Financing Responsiveness to Assessed Need during the Grand Bargain Period: Replication Dataset and Supporting Materials
Description
Replication package accompanying the study Humanitarian Financing Responsiveness to Assessed Need during the Grand Bargain Period. This package contains the analytical datasets, R scripts, computational environment documentation, replication instructions, validation materials, and supporting documentation required to reproduce the reported statistical analyses, tables, figures, and principal findings. The replication workflow has been independently tested and documented. Researchers following the supplied analytical workflow should reproduce the reported results, subject only to negligible computational precision or documented software-version differences.
Files
Steps to reproduce
Steps to Reproduce This replication package supports direct computational replication of the analyses reported in the associated study. Researchers should use the supplied analytical datasets, documented computational environment, and accompanying R scripts without modifying the analytical workflow, variable definitions, or model specifications. Download and extract the complete replication package. Preserve the folder structure, including 01_Data, 02_R_Scripts, 03_Reference_Replication_Results, 04_Validation, 05_Documentation, and outputs. Review the README, Replication Instructions, Computational Environment, and Data Dictionary before executing the analyses. Install R and verify that the computational environment is consistent with the accompanying documentation. The independent replication verification was performed using R 4.5.2 on macOS Tahoe 26.3.1 (aarch64-apple-darwin20). Minor software-version differences should not materially affect the reported results. Open a terminal (or RStudio), navigate to the root directory of the extracted replication package, and execute: Rscript 02_R_Scripts/00_run_all.R The workflow uses project-relative file paths and should be executed from the package root. The workflow automatically validates the analytical datasets, executes the descriptive and regression analyses, generates tables, figures, diagnostic outputs, confidence intervals, validation reports, and records the computational environment in sessionInfo.txt. Compare the generated outputs with the supplied Reference Replication Results. The generated file outputs/replication_check.csv provides a direct comparison between the reproduced analytical results and the reference values included in this repository. Researchers following the documented workflow should reproduce the reported statistical models, descriptive statistics, tables, figures, and principal findings, subject only to negligible computational precision or documented software-version differences. This repository supports direct computational replication of the published study. Analyses using modified datasets, alternative data-processing procedures, different model specifications, additional explanatory variables, or alternative computational environments constitute independent reanalyses rather than direct replications. Such analyses are valuable scientific contributions but should not be interpreted as tests of the computational reproducibility of the published analytical workflow.