Aggregating validity indicators embedded in Conners' CPT-II outperforms individual cutoffs at separating valid from invalid performance in adults with traumatic brain injury.
OBJECTIVE To study the effect of administration sequence on Conner's continuous performance test (CPT-II) scores in clients requesting psychological assessment. It was hypothesized that when administered at the end rather than beginning of a test battery, the test scores will show higher symptom severity. If present, order effects may cause the over- or underdiagnosing of ADHD. METHOD Participants were recruited at a Midwestern university's training clinic (16 men, 9 women; mean age = 22.4, SD = 10.2). The CPT-II was administered twice to each client: once at the beginning of the testing session and once at the end of their appointment. The clients completed at least a full Wechsler intelligence battery in between the CPT-II administrations. RESULTS Clients' ADHD index score (interpreted as percent confidence in an ADHD diagnosis) is more impaired at Time 2 (M = 53.3, SD = 29.0) compared to Time 1 (M = 39.4, SD = 22.5): t (24) = 3.93, p < .05, Cohen's d = .79. The number of T-scores above 60 on the subscales also changed from Time 1 (M = 1.92, SD = 1.73) to Time 2 (M = 3.12, SD = 2.05): t(24) = 3.47, p < .01, Cohen's d = .71. CONCLUSIONS If the CPT-II is administered later in a sequence of tests, it is more likely to yield scores in the impaired range. Order effects are more pronounced in individuals diagnosed with ADHD. Recommendations include the adoption of a standardized administration sequence, further research to investigate the nature of order effects, and a strategic use of order effects in ADHD assessment.