Current
What we can say today
Lux has versioned scoring, defined report rules and source-bank internal-consistency evidence. These are documented foundations, not completed Lux-specific validation.
Lux evidence
Ambitious about people. Honest about proof.
Lux is moving through an active, staged evidence-development programme. We publish what current evidence supports, state the limits, and strengthen claims only when pre-defined studies produce supporting results.
Current evidence status
Three plain answers.
Active
What we are studying
The evidence programme includes reliability, validity, fairness and utility questions, staged as suitable data and study designs become available.
Claim limit
What we do not claim
Lux is not clinically validated, diagnostic, predictive or a basis for personnel selection. Participant volume alone does not change those limits.
What we know
A documented starting point.
The active programme combines ongoing testing of versioned scoring and committed fixtures with staged Lux-specific empirical evaluation. Technical checks make the scoring pathway more inspectable; they do not by themselves establish psychological reliability or validity.
The source/reference evidence and technical architecture are useful context, but neither should be mistaken for completed Lux-specific validation.
Source-bank internal consistency
α .800–.898
The full ten-item Big Five item banks have documented internal consistency of α .800–.898 in a 603,322-response source/reference dataset.
The limit. This is source/reference internal-consistency evidence, not a count of Lux participants and not proof that every Lux output is valid. It does not establish that a bank measures one underlying construct, that Lux’s proprietary fields are valid, that Lux itself is reliable, or that Lux is fit for a particular use.
Lux also has an evidence-governance architecture: defined fields, versioned scoring, report eligibility and explicit claim limits. That makes the work more inspectable; it does not by itself establish psychological validity.
Evidence roadmap
A staged path, with status stated plainly.
No stage is treated as complete before its evidence supports that status.
-
Status: Current
Foundations documented
Versioned scoring, defined report rules, evidence boundaries and source/reference evidence provide an inspectable starting point.
-
Status: Active
Technical checks active
Ongoing checks of versioned scoring and committed fixtures make the scoring pathway more inspectable; they do not establish psychological reliability or validity.
-
Status: In progress / future, as described
Lux-specific empirical studies
Test–retest work is part of the active programme. Other staged studies proceed as suitable participant data and study designs become available.
-
Status: Future evidence gate
Stronger claims only after supporting results
Claims advance only when pre-defined studies produce results that support them, not when participation volume reaches a milestone.
What we are testing
The study questions, and where each stands.
Present claim limits
- Lux remains a descriptive assessment. It is not clinically validated, diagnostic, predictive or a basis for personnel selection.
- The programme is active, but it has not yet produced completed Lux-specific validation results.
-
Active
Versioned scoring tests
Ongoing testing of versioned scoring and committed fixtures, to detect unintended computational changes. This makes the scoring pathway more inspectable; it does not by itself establish reliability or validity.
-
Active
Test–retest reliability
Part of the active evidence programme. It will assess score stability as repeat-response data becomes available.
-
Staged
Convergent and discriminant validity
For proprietary Lux fields.
-
Staged
Representative norms
-
Staged
Intended-population fairness
-
Staged
Clinical utility
Staged studies will examine these questions as suitable data and study designs become available.
How claims advance
Evidence grows by study, not volume alone.
Eligible participant data enables pre-defined studies; participant volume alone does not establish validity, reliability, norms, fairness or clinical utility. Stronger public claims will follow only when study results support them.
- Eligible data
- Pre-defined study
- Supporting results
- Stronger claim
One example: the 500-response gate
Lux is a fixed-form assessment in its current phase. Lux-specific adaptive-calibration work will not begin before at least 500 eligible Lux responses are available. That gate has not yet been met.
Reaching it makes analysis possible. It does not establish validity, fairness, norms or utility.
Why calibration waits for 500 responsesHow we publish
Claims follow evidence. Never the other way.
Publish what current evidence supports.
State the limits beside each claim.
Strengthen claims only when pre-defined studies produce supporting results.
This page is the source of record for those boundaries. It will change as the programme produces findings that can support new claims.
Check our accountability
Australian registered charity, DGR endorsed, with public annual reports and named governance.
Research and support
Help the evidence advance responsibly.
Research partners can help answer defined questions. Explore reliability, validity, fairness, norms or utility questions with clear evidence gates and governance.