Across ten measures, a third of healthy adults have a low score.
31.6%, in a study of 5,046 cognitively normal adults, had at least one score at or below 1.5 SD. Give somebody twenty tests and a low score is close to expected. The commonest way to over-read a profile is to notice one.
Why a fixed rule cannot work
“Two scores below 1.5 SD” is a rule that means something different on a six-measure battery than on a twenty-five measure one. Every additional measure is another chance for a healthy person to fall below the line, so the number of low scores that should raise an eyebrow rises with the size of the battery.
At the one-SD line this gets worse rather than better: once a battery reaches any real size, the great majority of healthy people have at least one score below it. Counting to a number that everybody reaches is not counting.
The line Casebook uses, and why
| Measures in the battery | Low scores before the pattern is unusual |
|---|---|
| Up to 2 | 1 |
| 3 to 9 | 2 |
| 10 or more | 3 |
“Unusual” here means rarer than about one healthy person in ten. Low is 1.5 SD rather than 1.0, for the reason above. The point of the table is its shape: what counts as unusual depends on how many measures were given, and the rule has to say so.
Casebook counts rather than flags. The case page shows how many scores sit at or below the line and how many a battery of that size produces in people with nothing wrong — the two numbers side by side, so the comparison is the reader’s to make.
The other end: scores that have hit the floor
A Wechsler full scale bottoms out at 40, which is exactly 4 SD below the mean, and a raw score of zero still returns a scaled score of 1. So a floored score tells you the ability was below the range of the test, and nothing about how far below.
Profile shapes built over floored scores are the instrument’s shape rather than the person’s. When composite scores in a battery of three or more sit at or beyond 3 SD, the low-score reading says how many, and that they record the bottom of the test rather than the bottom of the person.
Where these numbers come from
- Oltra-Cucarella et al. (2021) — base rates of low scores in 5,046 cognitively normal adults
- Cook, Iverson & Karr (2025) — the same question in 1,269 children aged 7 to 17
- Binder, Iverson & Brooks (2009) — on abnormal scores being psychometrically normal
A low score is not a finding, and a scattered handful of them is not a pattern. Both of those sentences are load-bearing.