Use when
- A page interprets an endpoint as anxiety, pain, memory, motivation, or motor impairment.
- Multiple methods or controls are needed to support a construct.
- The claim depends on species, model, or apparatus assumptions.
Scientific Validation
Validation
Construct validity
Construct validity asks whether a method and endpoint support the biological interpretation being made.
Decision summary
Use construct-validity language when a page needs to separate the measured endpoint from the broader biological construct. This is the guardrail against turning one behavioral score into an unsupported disease or trait claim.
| Measured endpoint | Observed score, latency, threshold, count, or percentage. |
|---|---|
| Construct | The biological or psychological interpretation inferred from the endpoint. |
| Evidence need | Converging methods, controls, model rationale, and alternative explanations. |
| Failure mode | Endpoint treated as a direct trait without caveats or controls. |
Do not use when
- The page only defines a raw endpoint without interpretation.
- Evidence is too thin to make the construct claim; use a caveat or hold the page.
Caveats
- Construct validity is not created by a familiar method name.
- A single endpoint can support different interpretations in different contexts.
- Negative controls and alternative explanations matter as much as positive findings.
Reporting checklist
- Name the endpoint and construct separately.
- State what alternative explanations were checked.
- Report controls that support the interpretation.
- Avoid direct clinical or trait language when the method supports only a proxy readout.
- Link to validation evidence or keep the claim qualified.
References
Belzung and Lemoine 2011. Criteria of validity for animal models of psychiatric disorders.
Review of validity criteria including construct validity in animal models.
https://link.springer.com/article/10.1186/2045-5380-1-9
Willner 1984. The validity of animal models of depression.
Classic discussion of validity concepts in animal models.
https://pubmed.ncbi.nlm.nih.gov/6429692/
Related surfaces
Use these related surfaces to move from the scientific method question to the relevant product page, endpoint definition, analysis tool, or adjacent guide.
