ADHD glossary · Family perspectives
Outlier: investigate an unusual value before explaining it away
NIST's statistical handbook describes an outlier as an observation unusually far from other values in a random sample, with the meaning of “unusual” requiring context. [1]
General education and practical perspectives. Editorial standards.
For a reader, the practical question is what the value represents and how the report treated it. We are discussing a statistical observation, not labeling a person. The invented example below is not a formal detection test or an analysis of ADHD data.
The box with ninety items
Imagine a hobby group recording how many wooden tokens are in five boxes. Its handwritten note lists 8, 9, 10, 11 and 90. Ninety stands apart from the other recorded counts. That is a reason to ask a question, not enough information to decide that someone made a mistake.
One member wonders whether 90 was meant to be 9. Another asks whether the last line describes a storage carton rather than a small box. A third thinks the larger count may be correct. They leave all three possibilities as questions while checking the original record.
In this invented scene, the group has not yet found the answer. It would be misleading to replace 90 with 9 merely because the revised list looks tidier. It would also be misleading to invent a special explanation for the value without checking.
What can the actual report tell you?
Value being discussed:
What was measured and in which unit:
Source record or table location:
Reason the report calls it unusual:
Whether a correction or exclusion was made:
Explanation and method the authors provide:
What I have checked myself:
What is still only a possible explanation:
Our card is a reader's note, not instructions for applying a statistical threshold. For a real analysis, look for its actual measurement context and stated method; do not substitute a rule inferred from five invented numbers.
Do not confuse unusual with missing
The token note contains an observed entry of 90. If the fifth box had not been counted, that would be an unanswered measurement instead. Writing zero for an uncounted box would introduce a different assertion: that the box had no tokens. Keep those states distinct in this example.
An average or median is another question again: it summarizes a specified set. Before repeating a summary, look for whether the unusual entry was included and whether an exclusion was explained. Do not assume a published number includes every observation solely because the article mentions a larger initial count.
Keep the language attached to the data
A teen checking the note can ask about its unit without being made responsible for the group's whole record. A parent can admit an entry is unresolved. A sibling's different preference is not a statistical outlier simply because others disagree with it.
When an ADHD-related article discusses outliers, ask what measurement and method it means. This glossary does not identify anyone as abnormal, interpret clinical test results or decide which participants belong in a study.
Source
[1] National Institute of Standards and Technology. What are outliers in the data?. NIST/SEMATECH e-Handbook of Statistical Methods, section 7.1.6. Page-specific publication date not stated; accessed September 6, 2026. Opening definition and its context qualifier support [1]. The source's 90-observation dataset, box-plot rules and figures are not reproduced or taught here.
Related guides
Sources and what they support
These sources support the specific statements identified below. Our practical exercises are suggestions to consider, not evidence of a treatment effect.
-
7.1.6. What are outliers in the data? — National Institute of Standards and Technology
Source type: institutional.
Accessed .
Source check, September 6, 2026: Scoped source check. Passage: Definition and analyst/context qualifier at retrieved line7; final investigation paragraph line30 inspected as context. Limits: Context-dependent definition only; no prescribed detection threshold, box-plot rules, dataset, statistical test, deletion rule or clinical interpretation.
Supports: NIST's statistical handbook describes an outlier as an observation unusually far from other values in a random sample, with the meaning of “unusual” requiring context.
Correction check: Original full HTML was retrieved and the adopted passage, context and available date/footer information inspected. No notice changing that definition found there. Not an exhaustive history or correction/retraction database clearance.
Updates and corrections
Published .
Found a factual error or something unclear? Use the contact route on our Trust page. Read our research method for the evidence approach.