Authorised theGrowSpace clinicians only.
Each case links the child's screener submission(s) with the clinician-entered criterion measures and diagnosis. Click a case to open it.
| Case (child code) | School | Screeners | ASD | ADHD | SLD | Anx | Str | Criterion | Diagnosis |
|---|
Demographics + provenance for analysis. Retrospective (report-only) cases carry their age here since there is no screener submission.
Enter the standardised scores administered in the comprehensive assessment. These are what the screener profiles are validated against. Leave blank if not administered.
The clinical conclusion per condition. This is what sensitivity/specificity are computed against.
Live analyses over the linked cases. These stabilise as paired data accumulates — small n gives unstable estimates. Definitive reporting still belongs in R/Python/SPSS on the exported dataset.
Does the screener profile move with an established measure of the same construct? (Pearson r)
Does the screener separate children who do vs don't receive the diagnosis?
How well does the screener score classify the diagnosis? ROC curve, AUC, and a candidate cut-score (Youden's J). Higher screener % is treated as predicting a positive diagnosis.