Situational barriers reach far more people than permanent ones

What it is

The census slice of people with a disability determination is a thin line. The people who, in the same week, hold a child in one arm, cannot read a screen in sun, or open a video where sound must stay off, are a thick one. Situational impairment prevalence is a coverage claim, not a moral ranking: among people blocked by an interface at any moment, most have no diagnostic label and have not registered themselves as disabled users. Scheduling accessibility work from determination counts measures the most common failures with the narrowest counter.

Why it happens

Permanent impairment is a low, relatively stable base rate. Situational triggers are high-frequency environmental states — commuting, caregiving, outdoors, a device that has just failed — that almost everyone walks into repeatedly. The second layer is that measurement hides the thick line: people in a bind rarely turn on assistive technology, rarely hit “accessibility feedback,” and rarely flag disability on an account. Failure shows up as bounce, switching services, or never starting. Analytics then sees only the thin slice of permanent users who enabled captions or high contrast, and files the problem as niche.

Studying it

Compare three scales: official disability statistics; diaries or intercepts after a failed task that ask whether the environment made a channel unusable; and opt-in rates for assistive technology or accessibility features. Treat diagrams such as Microsoft Inclusive 9 here as a scale discussion tool, not as a tutorial in user modelling.

Independent variables: counting frame (determination population / situational triggers that week / feature enablement). Dependent variables: order-of-magnitude gaps across the three frames; share of sessions labelled “accessibility users” in analytics; share of incomplete tasks that self-report an environmental limit.

Do not read feature enablement as situational scale — most situational users never go looking for that switch.

Where it stops holding

A larger count does not automatically outrank a smaller permanent need that has no alternative path; using the majority to cut “niche” assistive technology turns the coverage argument backwards. Peak concurrent situational load is also not mean annual exposure — report the peak and the mean separately. Some situations (weeks after surgery) sit closer to temporary; folding them into “everyone has been there” inflates everyday coverage. Internal specialist tools may almost never see those commute situations, so the coverage argument needs a different frame.

Applying it

• Planning decks may not show determination percentages alone; the same page must give a rough order of magnitude for “people who could hit this failure this week without a diagnosis.”
• Usage data for accessibility features should break out the share of people who use them without a disability flag, so enablement is not read as the permanent-user headcount.
• Do not let the large situational count justify a botch that only holds for a few minutes — coverage answers who will hit the wall, not how roughly the work may be done.
• How to check: pick a live failure (contrast in sun, a gesture that cannot be finished one-handed, a video that cannot be watched muted). Ask via support, diary, or intercept “did the environment stop you,” not “are you disabled.” If environment answers dominate, the determination frame has shrunk coverage.

Related

• Same group: J1.08.1 Permanent, temporary, and situational barriers overlap in need · J1.08.3 One accommodation can serve all three groups
• Nearby: A11.08 Situational impairments · J1.05 Inclusive Design Principles
• Search terms: situational impairment prevalence · persona spectrum · coverage vs census