The French (Canada) and Spanish (North America) versions of the CAARS 2 were developed using best practices, as outlined within this chapter. The psychometric evidence compiled within gives solid evidence that the linguistic and culturally sensitive translation produced versions of the CAARS 2 that are both valid and reliable.

Reliability. Evidence of reliability was established through the following:

• Excellent internal consistency of the translated forms (median omega coefficients = .94 and .96 for the French version of the CAARS 2 Self-Report and Observer, respectively; .93 and .97 for the Spanish Self-Report and Observer form, respectively), which was comparable to the results found in the English version in the Normative Samples.
  
  
• High degrees of test information, which met common benchmarks for high measurement precision.
  
  

Validity. Evidence of validity was established through the following:

• Measurement invariance of the Content Scales factor structure, which was established across all translated forms and their English counterparts.
  
  
• High correlations between Content Scale scores for translated forms and their English counterparts (median corrected r = .91 and .89 between the French and English CAARS 2 Self-Report and Observer, respectively; .93 and .84 between the Spanish and English CAARS 2 Self-Report and Observer, respectively).
  
  
• Negligible differences between the average Content Scale scores for translated forms and their English counterparts (maximum Cohen’s d for French vs. English: Self-Report = 0.14 and Observer = 0.13; for Spanish vs. English: Self-Report = 0.19 and Observer = 0.13).
  
  
• No significant difference between CAARS™ 2–ADHD Index probability scores across the translated forms and their English counterparts (effect size r: French Self-Report = -.10, French Observer = -.08, Spanish Self-Report = -.01, Spanish Observer = -.02).
  
  
• A high degree of concordance in classifications for the Associated Clinical Concern Items (median for French Self-Report = 93.6% and Observer = 92.8%; Spanish Self-Report = 93.8% and Observer = 93%), Impairment & Functional Outcome Items (median for French Self-Report = 92% and Observer = 91.8%; Spanish Self-Report = 92.6% and Observer = 90%), Negative Impression Index (French Self-Report = 92.7% and Observer = 97.9%; Spanish Self-Report = 93.3% and Observer = 92.6%), and Inconsistency Index (French Self-Report = 93.1% and Observer = 91.8%; Spanish Self-Report = 92.9% and Observer = 93%).