How Older and Younger Adults Differ in Their Approach to Problem Solving on a Complex Website
Peter G. Fairweather · 2008 · Proceedings of the 10th International ACM SIGACCESS Conference on Computers and Accessibility (Assets '08) · doi:10.1145/1414471.1414485
Summary
This IBM Research paper challenges the prevailing assumption that older adults use the web in the same way as younger adults, just more slowly and less accurately. The study examined how 28 participants aged 18-73 navigated a complex newspaper website to solve a realistic problem-solving task: finding employment opportunities for a friend or relative within 75 miles of the participant's location. The task required participants to categorize jobs, use domain vocabulary (e.g., "ward clerk," "chart," "logistics"), and navigate multiple search pathways. The researcher used a proxy server to log every page visit and constructed navigation graphs for each participant. These graphs were combined into a single meta-graph, which was then algorithmically partitioned into six "zones" representing distinct neighborhoods of activity within the website — including an all-topic search zone, a combination-direct search path, a keyword-directed zone, a guided-incremental search zone, a site-exit zone, and an empty-set zone (fruitless searches).
Key findings
Neither age nor experience significantly predicted task success (46.4% completion rate overall), but session elapsed time was significantly affected by both age and experience as covariates (F(1,26) = 5.317, p < .05). The key finding was strong age-related correlations with visits to particular website zones. Older adults were significantly drawn to the "guided-incremental" search zone (r = .522, p < .01), which offered a low-risk, step-by-step approach that channeled users through small navigational increments toward a solution. Older adults also visited the "site exit" zone significantly more often (r = .459, p < .05) and the "empty set" zone more often (r = .342, p < .05). Experience had a significant negative correlation with the guided-incremental zone (r = -.343, p < .05), meaning more experienced users actively avoided it. Paradoxically, the guided-incremental zone had a strong negative correlation with task success (r = -0.506, p < .01) — the path that felt safest actually led to more failures because early missteps in categorization compounded through the incremental process, leading users far from their goal.
Relevance
This study has important implications for accessible web design, particularly for aging populations. It demonstrates that older adults don't simply do the same things more slowly — they systematically choose different strategies and pathways, gravitating toward approaches that minimize cognitive load but may paradoxically reduce success. For web designers, this means that "guided" step-by-step interfaces, while seemingly helpful, can trap older users into compounding errors. The authors recommend providing feedback about progress toward goals across entire sessions, not just individual actions, and making it easier for users to recover from early missteps. The finding that experience moderates age-related tendencies suggests that design interventions, not just age-related accommodations, can meaningfully improve outcomes. This work cautions against user-data-based evaluation methods that assume all users follow similar paths.
Tags: aging · web navigation · cognitive accessibility · usability · older adults · problem solving · web design · user behavior · cognitive load