DATA in "Beyond Formal Disclosure: Information Paths and Evaluation Responsibility in European E-Procurement"
Description
This dataset accompanies the manuscript “Beyond Formal Disclosure: Information Paths and Evaluation Responsibility in European E-Procurement.” It provides the nonpersonal aggregate reporting data and code-inspection materials supporting the reported tables, figures, and diagnostic summaries. The deposit contains unrounded table and figure source data; final aggregate reporting outputs; source code for the rank-free overlap-weighted estimator, reporting builder, and R figures; standalone synthetic test fixtures and public test scripts; design chronology records; environment information; checksums; a README; and a data dictionary. The deposit supports inspection of aggregate results, re-rendering of Figures 1 and 2, checksum verification, and execution of two standalone synthetic component test suites. It does not contain the 439 MB raw TED XML collection, personal or contact fields, cookies or credentials, exact notice identifiers, row-level frozen analysis inputs, linkage records, or production weights. It is therefore not a raw-to-result replication archive and cannot independently recompute the empirical endpoint estimates. Authoritative procurement notices remain available through Tenders Electronic Daily (TED) under the applicable European Union reuse terms. Original project code is licensed under the MIT License. Original publicly shareable aggregate materials are licensed under the Creative Commons Attribution 4.0 International License, to the extent that the authors hold the relevant rights. These licences do not cover TED source records, third-party content, or third-party software. No sensitive personal data or ethics-restricted data are included. See README.md, DATA_DICTIONARY.md, LICENSE_STATUS.txt, the detailed licence files, THIRD_PARTY_NOTICE.txt, and checksums/SHA256SUMS.txt.
Files
Steps to reproduce
STEPS TO REPRODUCE 1. Download and extract JRT_Mendeley_Data_Archive.zip, preserving the top-level directory. 2. Read README.md and DATA_DICTIONARY.md for the archive scope, file map, variables, limitations, and licensing. 3. From the extracted archive root, verify file integrity with `shasum -a 256 -c checksums/SHA256SUMS.txt` on macOS or `sha256sum -c checksums/SHA256SUMS.txt` on Linux. 4. Create an isolated Python environment and install the listed dependencies with `python3 -m pip install -r requirements.txt`. 5. Run the supported public synthetic test entry point with `sh run_public_tests.sh`. The two final `OK` summaries indicate success. Some successful negative command-line tests deliberately print `error: unrecognized arguments` or `BLOCKED_NOT_READY`. 6. Re-render Figure 1 with `Rscript code/figures/figure_01_ddf.R`. 7. Re-render Figure 2 with `Rscript code/figures/figure_02_target_sensitivity_revision.R source_data/figures/figure_02_target_sensitivity_revision.csv figure2_rebuild`. The public tests use synthetic fixtures only. They do not reconstruct endpoint membership, linkage, weights, or empirical estimates from raw TED records. The archive’s reproducibility claim is limited to checksum verification, aggregate-result inspection, figure rendering with the listed R packages, and the two standalone synthetic test suites.
Institutions
- The University of OsakaOsaka, Osaka