Web accessibility in Spanish city councils: a challenge for the democratic inclusion and well-being of citizens

The study situates web accessibility within the broader context of digital democracy, where ICTs are expected to enhance transparency, participation, inclusion, and overall well-being. While governments are rapidly digitising services, risks arise when a digital-by-default approach overlooks those lacking access or skills. A gap is identified in linking digital democratisation with standardised validation of web accessibility. The research question is: Do local councils take international web accessibility guidelines into account to enable a participatory and inclusive digital environment for citizens? The focus is on Spanish municipal portals, a critical scale for citizen interaction, with attention to vulnerable groups such as people with disabilities and older adults. The study examines whether local councils have adapted to evolving WCAG standards and how this adaptation has evolved over time.

The paper reviews work on digital democracy and inclusion (e.g., Becker, Kneuer and Datts), highlighting how digital transformation can foster participation but may also generate barriers. It underscores international frameworks and obligations such as the UN CRPD and Spain’s Digital Agenda, and the W3C’s WCAG as the global benchmark for accessibility. Prior accessibility studies often focus on universities or national portals, with fewer on local councils; an early Spanish study by Discapnet (2008) evaluated municipal portals under WCAG 1.0. Accessibility benefits extend beyond users with disabilities to those with slow connections and older adults, per W3C. Recent research emphasises responsive and inclusive design, yet evidence shows persistent issues such as missing alt text and unclear links. The paper notes a lack of longitudinal, standardised evaluations of municipal websites using WCAG criteria and a need to integrate inclusion as a social challenge within e-democracy research.

Design: Exploratory, quantitative, longitudinal analysis of municipal website accessibility using automated evaluation complemented by manual review of declared compliance. Sample: Stratified random sample of 18 Spanish provincial-capital city councils, grouped by city size and per capita income into four groups (metropolitan; large; medium; small). For each portal, five page types were selected: (1) Home page; (2) Standard internal page; (3) Page with tables; (4) Page with a form; (5) Search results page (keyword: “web accessibility”). Due to missing page types on some portals, 88 pages were analysed in 2015 and 87 in 2021 instead of the planned 90. Period: Evaluations conducted in 2015 and revisited in 2021 (March–May) to assess evolution and compliance with WCAG 2.0/2.1. Instrument: WAVE Web Accessibility Evaluation Tool (WebAIM) via the Chrome extension, recommended by W3C. Indicators recorded included: Errors (violations of WCAG 2.0), Alerts (potential issues needing manual verification), Features (identified accessibility implementations), Structural elements, HTML5/ARIA usage, and Contrast errors (text-background color contrast problems). An observatory-style manual check reviewed each portal’s published accessibility declaration (e.g., WCAG 1.0/2.0/2.1, level A/AA/AAA, or unspecified). Standards: WCAG 1.0, 2.0, 2.1 levels A, AA, AAA; assessed primarily against WCAG 2.0 criteria while noting evolution toward 2.1 and forthcoming 2.2/3.0.

- Total errors decreased by about 17%, from 473 (2015) to 390 (2021), indicating only a slight improvement over six years. - Approximately 39% of portals increased their total errors in 2021 compared to 2015, notably in Groups 1 and 4. - Alerts increased by more than 32% overall from 2015 to 2021; about 72% of portals had more alerts in 2021 than in 2015. - Contrast errors increased by about 45%; roughly 50% of portals had higher contrast errors in 2021, particularly in Groups 3 and 4 (Group 1 was an exception). - Accessibility “Features” and “Structurals” detected by WAVE increased by ~24% and ~83%, respectively, and HTML5/ARIA indicators more than doubled, showing technological evolution but not necessarily compliance. - Common recurrent errors (mostly WCAG 2.0 Level A): 1.1.1 Non-text Content (missing alt text for images); 2.4.4 Link purpose (in context) and empty links/buttons lacking text values—issues that confuse screen reader users. - By page type: Group 1 showed worsening across all page types, with forms having the most errors in both 2015 and 2021; Group 2 improved across page types except for homepages; Group 3 had higher error counts overall with slight improvement on standard and search pages; Group 4 worsened across all page types. - Declared compliance: In 2015, ~67% claimed WCAG 1.0 AA. By 2021, ~21% still claimed WCAG 1.0 AA, while more portals claimed WCAG 2.0 AA/AAA; 33.33% did not specify any level in 2021. Only Santa Cruz de Tenerife claimed WCAG 2.1 AA. - No clear relationship between city size/budget and accessibility performance; some larger cities worsened, and improvements did not align with higher resources.

Findings indicate that Spanish municipal portals have not sufficiently integrated WCAG into their digital transformation to ensure inclusive, participatory e-democracy. Despite a modest reduction in total errors, increases in alerts and contrast issues suggest growing or unresolved barriers for users, especially those relying on assistive technologies. The most prevalent violations (missing alt text, empty links/buttons) directly affect screen reader navigation and comprehension, undermining inclusion for people with disabilities and older adults. The divergence between increased technical features/structures and persistent or worsening violations underscores that adopting modern frameworks does not guarantee accessibility compliance. Declared compliance levels have nominally evolved from WCAG 1.0 to 2.0, yet many portals either do not specify their level or make claims not reflected in automated findings, implying limited institutional awareness and governance around accessibility. The absence of a correlation between city size/budget and accessibility performance suggests organisational priorities and processes, rather than resource levels alone, drive outcomes. Addressing the research question, local councils largely do not yet meet international accessibility guidelines sufficiently to support inclusive digital participation. The study emphasizes the need for systematic remediation, heuristic expert reviews, and user testing with people with diverse disabilities, alongside validation of assistive technologies, contrasts, captions, and text settings, to translate digitalisation into genuine inclusion.

Web accessibility remains a significant challenge for the analysed Spanish city councils. Transition efforts from WCAG 2.0 to 2.1 corresponded to only a small reduction in total errors, while alerts and contrast errors rose, indicating that barriers persist or have increased in key areas. Overall, the evolution from 2015 to 2021 does not demonstrate sufficient adaptation to inclusive design for e-democracy, despite some portals’ accessibility statements. Strengthening accessibility can enhance local-level participation, credibility, transparency, and trust in public institutions. Future work should include evaluating municipal mobile applications required to meet WCAG 2.1, assessing citizen participation mechanisms on municipal sites, and triangulating with user evaluations to provide a holistic perspective. Broader adoption of standardised WCAG-based methodologies over time can support objective comparisons and guide continuous improvement.

- Reliance on automated tool (WAVE) outputs requires complementary expert heuristic reviews and user testing with people with various disabilities and assistive technologies to validate findings. - Rapid evolution of web technologies and standards (e.g., forthcoming WCAG 2.2 and 3.0) may outpace evaluations and affect comparability. - Some planned page types were unavailable for certain portals, reducing the total number of pages analysed (88 in 2015; 87 in 2021), which may slightly affect comparability across portals and years.