Standardized. Calibrated. Built for Clinical Credibility.

OT Wizard transforms everyday clinical observations into defensible, standardized scores — using real therapist-rated performance data and Rasch psychometric modeling, the same scientific framework behind tools like the PEDI, SFA, and Vineland.

🔬 Our Scientific Foundation

Trusted Developmental Milestones
We start with established developmental research from respected sources, including:

• University of Washington Developmental Milestone Charts
  
  

• Established pediatric assessment standards
  
  

• Evidence-based developmental expectations
  
  

Advanced Statistical Analysis (Rasch Modeling)
Think of Rasch analysis like GPS recalibration. Just as your GPS uses real driving data to give better directions, we use real therapist-collected, de-identified data to refine developmental expectations. This ensures items are anchored in both research and real-world performance.

Benefits for Practice

• For Clinicians: Scientifically validated expectations provide confidence in identifying concerns, with efficiency gains from smart start/stop rules and rubrics.
  
  

• For Administrators: Results are psychometrically defensible, consistent across raters, and meet compliance standards.
  
  

• For the Field: Every de-identified evaluation contributes to advancing pediatric therapy knowledge and refining developmental expectations.
  
  

🧪 Validation Through Rasch Analysis (2025)

In 2025, over 300 therapist-scored evaluations from OT Wizard were analyzed using Rasch modeling. Results show:

High-Quality Item Functioning

• The 0–4 scale functions well and is used consistently across raters and items
  
  

• Most items fit the Rasch model, measuring a clear shared ability (e.g., functional independence)
  
  

• Reliability and separation indices indicate the tool can meaningfully differentiate skill levels
  
  

Clear Construct Measurement

• PCA confirmed scores are mostly unidimensional — reflecting a coherent developmental ability
  
  

• A small secondary dimension was flagged, guiding refinement
  
  

Differential Item Functioning (DIF)

• Items tested across age, gender, rater type, and geographic region
  
  

• A few items (e.g., Q1, Q4) showed DIF and are under review for fairness
  
  

Conclusion: OT Wizard is already psychometrically sound and actively improving through iterative data analysis and refinement.

🔢 How Items Are Scored

Unlike static tests or opinion-based checklists, OT Wizard uses performance-anchored rubrics applied by licensed OTs:

Each rating is transformed into a standardized, ordinal score that supports consistent comparisons over time and across students.

📊 Current Score Reporting

• Domain Scores: Scaled 0–100
  
  

• Composite Score: Scaled 0–1000
  
  

• Interpretation Bands: “Not Yet Observed”, “Beginning” “Emerging,” “Developing”,“Proficient,” “Mastered”
  
  

• Flagging: Scores flagged when deviating significantly from expected developmental age
  
  

All AI-generated text (summaries, goals) is based on therapist-collected scores — never raw child data, and never without clinician review.

🔒 Secure. Transparent. FERPA/HIPAA-Aligned.

• Therapist responses stored securely on AWS with BAA in place
  
  

• No PHI shared with AI services — only de-identified, structured data
  
  

• Detailed audit logs, score flagging, and user oversight built in
  
  

📈 Next Steps in Standardization

OT Wizard is progressing from defensible rubric-based scores toward a fully standardized, norm-referenced tool.

Age Bands

• 0–5 years: 6-month intervals (e.g., 24–30 mo, 30–36 mo)
  
  

• 6–11 years: 1-year intervals (e.g., 7:0–7:11)
  
  

• 12–18 years: 2-year intervals (e.g., 15–16, 17–18)
  
  

Sample Size Goal

• At least 250 data sets per age band
  
  

• Data contributed exclusively by licensed OTs through routine practice
  
  

Normative vs. Clinical Samples

• Norms anchored on children without OT intervention (general education population)
  
  

• Clinical data included to strengthen calibration, fairness testing, and clinical validity
  
  

Crowdsourced Data Collection

• Data collected through everyday OT practice in OT Wizard
  
  

• All contributions are de-identified, HIPAA/FERPA-compliant, and used solely for psychometric calibration
  
  

• Every evaluation contributes to advancing pediatric therapy science
  
  

Psychometric Transparency

• Annual Rasch calibration reports (item fit, PCA, DIF) shared with users
  
  

• Items revised and retested when misfit or DIF is flagged
  
  

Timeline

• End of 2025: First percentile tables for ages 3–5
  
  

• Q1–Q2 2026: Percentiles and standard scores extended through age 12
  
  

• Q3–Q4 2026: Full coverage through age 18, with annual recalibration thereafter
  
  

🎯 Guidance for Schools & Clinics

• Interpretation Framework: Scores link to functional descriptors (Needs Support, Emerging, Proficient, Advanced)
  
  

• Eligibility & Progress: Criterion-referenced today, with full norm-referencing available 2025–26
  
  

• Transparency Promise: Data is collected only by licensed OTs, securely stored, and shared in aggregate
  
  

📩 hello@otwizard.com