Italian Web Accessibility Study (IWAS): Automated WCAG Barrier Measurement Dataset for 10,022 Italian Domains

Abstract

Italian Web Accessibility Study (IWAS): Automated WCAG Barrier Measurement Dataset for 10,022 Italian Domains This dataset accompanies the Italian Web Accessibility Study (IWAS), a large-scale, reproducible observational study of automatically detectable accessibility barriers across the Italian web. The study examines 10,022 .it domains drawn from the Italian portion of the Tranco Top 1 Million ranking (list 46XZX, dated July 8, 2026). This is the same sampling frame used in the previously published Italian Web Security Study (IWSS), enabling future matched-domain comparisons between web security and accessibility. Data were collected between October 9 and October 13, 2026, using Playwright with Chromium and axe-core 4.14.0, following a frozen version 1.0 protocol. Up to five representative pages per domain were selected through deterministic homepage, functional-page, content-page, deep-path and seeded-page sampling. Dataset overview 10,022 domains in the original sampling frame. 7,573 domains successfully evaluated (75.6%). 2,449 domains with unsuccessful evaluations (24.4%). 36,038 sampled-page records. 34,641 successfully evaluated pages. 7,328 evaluable domains with at least one automatically detected accessibility violation (96.8%). 106,815 violation-rule occurrences. 890,680 affected DOM nodes. The most frequently detected accessibility issues concern insufficient colour contrast, links without accessible names, insufficient interactive target sizes, and missing or inadequate image alternatives. The released dataset includes domain-level measurements, page-level measurements, automatically detected accessibility violations, incomplete checks requiring further review, quality-control summaries, and methodological documentation. Data protection and reproducibility Domain identifiers are pseudonymized using unsalted SHA-256 hashing, compatible with the identifier scheme documented for the Italian Web Security Study (IWSS). URL identifiers are also hashed. Raw page URLs, page titles, DOM selectors, and HTML content are excluded from the public tabular export. These identifiers are pseudonymized rather than strictly anonymous, because dictionary matching against publicly known domains remains possible. The dataset includes unsuccessful measurements to support transparent evaluation of nonresponse and measurement bias. The acquisition methodology, software environment, and operational recovery procedures are documented. Interpretation and limitations Automatically detected axe-core violations must not be interpreted as comprehensive WCAG conformance determinations. The evaluation does not replace manual accessibility testing, keyboard interaction testing, assistive technology assessment, or complete website conformance evaluation. The original Tranco sampling frame is popularity-ranked and is not statistically representative of every Italian website. Non-evaluable domains must not be classified as either accessible or inaccessible. Related research Associated scientific paper: Manetti, G. (2026). Italian Web Accessibility 2026: An Observational Analysis of 10,022 Domains and 34,641 Web Pages. PerseoDesign Research. https://doi.org/10.5281/zenodo.23347092 Related security dataset: Manetti, G. (2026). Italian Web Security Study (IWSS): Passive Security Measurement Dataset. https://doi.org/10.5281/zenodo.21323346 The IWAS dataset is distributed under the Creative Commons Attribution 4.0 International license (CC BY 4.0).