Related Experiment Videos
A multitrait-multisource confirmatory factor analytic approach to the construct validity of ADHD rating scales
Rapson Gomez1, G Leonard Burns, James A Walsh
1School of Behavioural and Social Sciences and Humanities, University of Ballarat, Ballarat, Victoria, Australia.
Psychological Assessment
|April 5, 2003
Summary
This study found that parent and teacher ratings of attention-deficit/hyperactivity disorder (ADHD) symptoms showed more variance due to the rater (source) than the actual ADHD traits. This suggests limitations in the construct validity of ADHD rating scales.
Area of Science:
- Psychology
- Child Psychiatry
- Psychometrics
Background:
- Attention-deficit/hyperactivity disorder (ADHD) is a common neurodevelopmental disorder.
- Accurate assessment of ADHD symptoms is crucial for diagnosis and intervention.
- Rating scales are widely used but their construct validity requires ongoing evaluation.
Purpose of the Study:
- To evaluate the construct validity of ADHD rating scales using a multitrait-multisource (MTMS) design.
- To examine the influence of trait (inattention, hyperactivity/impulsivity) and source (parent, teacher) factors on ADHD symptom ratings.
Main Methods:
- Confirmatory factor analysis was employed to model the MTMS design.
- Data were collected from two independent samples of elementary school children in Australia (N=1,475) and Brazil (N=285).
- Parents and teachers rated children's ADHD symptoms.
Main Results:
- Most ADHD symptoms exhibited greater variance attributed to the source (rater) than to the trait (ADHD dimensions).
- This pattern was consistent across both Australian and Brazilian samples.
- Evidence for convergent and discriminant validity of ADHD symptoms measured by rating scales was weak.
Conclusions:
- Strong source effects indicate that rater perspectives significantly influence ADHD symptom ratings.
- The findings raise concerns about the construct validity of current ADHD rating scales.
- Implications for the interpretation and use of ADHD rating scales in clinical and research settings are discussed.