Frontiers in Psychiatry (Oct 2020)
Reliability and Validity of the Simplified Chinese Version of the Aberrant Behavior Checklist in Chinese Autism Population
Abstract
Background: The Aberrant Behavior Checklist (ABC) is a widely used scale in autism clinical intervention research for the assessment of core symptoms and comorbid emotional and behavioral problems among people with autism. The aim of this study was to examine the psychometric properties of the Simplified Chinese version of the Aberrant Behavior Checklist (SC-ABC) using a sample of people with autism in a Chinese population.Methods: In total, we enrolled 799 patients aged 1.5–33 years old. We collected data using the SC-ABC (n = 799), Autism Behavior Checklist (n = 743), Attention Deficit Hyperactivity Disorder Rating Scale-IV (ADHD-RS-IV) (n = 433) and Achenbach Child Behavior Checklist (CBCL) (n = 319). Eighty-four patients were separately assessed with the SC-ABC by two caregivers simultaneously. Forty-four patients were assessed with the SC-ABC again by same caregiver 2 weeks after the first assessment. SC-ABC data from the whole sample were used for confirmatory factor analysis. We evaluated criterion validity using Spearman's correlation coefficient between scores of the SC-ABC and scores of the Autism Behavior Checklist, ADHD-RS-IV and CBCL separately in the whole sample and different age groups. We calculated the intragroup correlation coefficients and Spearman's correlation coefficient for interrater reliability in 84 samples and test-retest reliability in 44 samples. We conducted Cronbach's α for internal consistency.Results: For the SC-ABC, the intragroup correlation coefficients of five subscales and the total score in interrater and test-retest reliability ranged from 0.87 to 0.92 and from 0.93 to 0.97 (all P < 0.01). The Spearman's correlation coefficient of five subscales and the total score in interrater and test-retest reliability ranged from 0.78 to 0.85 and 0.86 to 0.94, respectively (all P < 0.01). Cronbach's α of five subscales and the total score ranged from 0.75 to 0.96 (all P < 0.01). The Spearman's correlation coefficient for criterion validity for the whole sample and different age groups ranged from 0.39 to 0.76 (all P < 0.01). The model fit for the original five factor model was acceptable, with fit indices of SMR = 0.062 and RMSEA = 0.052.Conclusions: The SC-ABC has satisfactory psychometric properties and can be used in the assessment of core symptoms and comorbid emotional and behavioral problems in patients with autism.
Keywords