Data, protocol and code for: Computing Imbalance Indices in Signed Networks: A Systematic Mapping and Computational Survey

This dataset is the supplementary material of the article “Computing Imbalance Indices in Signed Networks: A Systematic Mapping and Computational Survey” (Alonso-Diaz, Riquelme and Olivares). It documents every step of the review so that the selection, coding and analyses can be audited and reproduced. The review covers the literature on the frustration index and the correlation clustering index in signed networks. The search was run in Scopus and Web of Science on 15 September 2026 and complemented with Google Scholar. It identified 767 unique records, which led to 133 included reports and 131 primary studies, reported following PRISMA 2020. Contents: Review protocol: research questions, search strings, eligibility criteria, selection, data extraction, reporting completeness index, venue classification, analysis and deviations. Deposited retrospectively, since the review was not registered before the searches. Screening log: the decision for each of the 767 records at title/abstract and full-text stages, with the criterion applied and the reason. Included reports and venues: the 133 included reports and the disciplinary classification of each study’s publication venue. Extraction and coding: the full data-extraction matrix for the four research questions, and the coded variables behind every count, table and statistical test in the article. Citation network: the internal citation network of the corpus (495 citations), with the textual evidence for each citation. Code: Python scripts that reproduce the statistical analyses (χ² tests with Monte Carlo p-values and Holm correction, Wilson intervals, Fisher test, sensitivity analyses, network homophily) and recompute the PRISMA flow counts. PRISMA 2020 checklist. The README file describes each file and how to run the scripts, and the codebook defines every variable. Data and documentation are released under CC BY 4.0 and code under the MIT License.