Automated multi-criterion assessment of university web presence: validating an indicator suite against student perceptions and admission outcomes across Ukrainian higher-education institutions

This deposit contains the full reproducibility package for the paper "Automated multi-criterion assessment of university web presence: validating an indicator suite against student perceptions and admission outcomes across Ukrainian higher-education institutions" (target journal: Computers & Education). The dataset comprises:- EDBO registry data — Metadata for 511 Ukrainian higher-education institutions sourced from the public EDBO API (registry.edbo.gov.ua), including institutional details, active study programmes, and admission outcomes.- Automated web-quality indicator suite — 25+ indicators measured across 268 reachable HEI websites using a deterministic three-source pipeline: static DOM heuristics (sitemap freshness, social proof, schema.org presence, navigation structure, mobile responsiveness), Lighthouse audits (Performance, Accessibility, Best Practices, SEO scores, plus six Core Web Vitals metrics), and axe-core WCAG 2.1 AA violation counts.- Student survey data — Anonymised responses from 84 students at Kryvyi Rih State Pedagogical University (September 2024) on perceived importance of web-quality dimensions and information-source preferences.- Crawl artifacts — 803 raw Lighthouse JSON reports and 256 axe-core JSON reports from three independent runs per HEI website.- Analysis outputs — Reliability coefficients, factor analysis scores, convergent validity correlations, negative-binomial GLM results, and moderation analysis outputs. All indicator extraction and analysis code is included (Python 3.12, Node 20, R 4.4), orchestrated via a Snakemake pipeline for full end-to-end reproducibility. The nine web-quality dimensions operationalised are: visual modernity, mobile responsiveness, applicant content, navigation clarity, calls to action, content freshness, social proof, branding, and web accessibility.