A single-case finding is not a general law until it has been replicated

What it is

One case can show that something is possible, and it can strongly challenge a claim that something never happens. Until independent observations repeat it, it is not a general law. Unreplicated single-case findings should be labeled as existence claims or propositions still being formed: this mechanism showed up here. Writing them as “users always…” or “this kind of interface necessarily…” lets one trajectory legislate a class of worlds.

Why it happens

A law claims stability: within stated conditions, the relation will appear again. One observation supplies no evidence of “again,” and cannot separate a stable mechanism from local coincidence or from biography and organization unique to that case. The power of a single case is depth: timelines, anomalies, and rival explanations can be pursued in detail. Depth does not convert into breadth. For a universal claim, one counterexample is enough to overthrow it; for an existence claim, one positive instance is enough. Most product decisions sit between those poles—is this typical, does it hold elsewhere?—which a single case cannot answer on its own.

Studying it

Record the logical type of the single-case claim: existence, rebuttal of a universal, or a proposition to be tested. When planning a repeat, define in advance what counts as the same phenomenon returning—the same failure pattern and condition structure, not the same complaint wording. Literal replication looks for nearby conditions; theoretical replication, if the proposition is to be strengthened, should change one condition claimed to be critical. With no second case, keep the modality as “was observed,” and list what should be seen next if the proposition is true.

Where it stops holding

Critical incidents, extreme users, and unique organizational arrangements often have to start as single cases; waiting is not a denial of their value. N-of-1 designs in clinical or safety work can be legitimate, but those are repeated measures on an individual, not one user standing in for a market. “We have all seen something like this” inside a team is not independent replication unless those observations are taken in as inspectable new cases.

Applying it

• Mark single-case items on the findings page as “existence / awaiting replication”; keep them out of universal sentences.
• If the case will drive a redesign, write the replication plan beside it: on which further people and tasks it must be seen again.
• Using it to kill “this cannot happen” is legitimate; using it to rank resources needs added frequency or a second site.
• In review, ask which part of the sentence would have to change if a contrary case appeared tomorrow; if none of it would, the sentence has already been written as a law.

Related

• Same group: Q4.10.1 A conclusion’s scope is bounded by the sample · Q4.10.2 A change of context can void a conclusion · Q4.10.3 Reports must state boundary conditions · Q4.10.4 Qualitative work seeks analytic generalization, not statistical representativeness · Q4.10.6 Recheck premises before transferring across cultures or platforms · Q4.10.7 Packaging small-sample findings as universal truths is a common overreach
• Adjacent: Q1.02 Exploratory and confirmatory work · Q1.09 Exploratory versus confirmatory studies
• Search terms: single-case finding · replication · existence proof