|
| Psychometric Issues in the Behavioral Treatment of Children With Autism |
| Sunday, May 30, 2010 |
| 10:30 AM–11:50 AM |
| 203AB (CC) |
| Area: AUT; Domain: Applied Behavior Analysis |
| Chair: Gerald E. Harris (Texas Young Autism Project) |
| Discussant: Gerald E. Harris (Texas Young Autism Project) |
| CE Instructor: Susan Ainsleigh, Ed.D. |
| Abstract: While not as exciting as new intervention techniques, accurate assessment of children with autism is an important and necessary conjunct of successful ABA treatment. The power and credibility of ABA in this area is demonstrated through measurement. The 3 presentations in this symposium present data that increases the psychometric knowledge, and thus the utility, of widely used measures of intelligence and behavior problems in the autistic population. Data were collected from comprehensive assessments of a large sample of children diagnosed with autistic disorder as they participated in behavioral treatment programs. Good sample sizes and appropriate data analytic procedures are important features of these studies. The first presentation looks at treatment outcome predictive value of the CBCL. The second data-based presentation describes extended normative tools for using the WPPSI-III with children with autism. The third presentation looks at how well an ABA intervention decreases cognitive variability in children. All three of these presentations represent another step forward in our ability to convincingly demonstrate the effectiveness of behavioral interventions. |
| |
| Predicting the Reduction of Positive Signs of Autism From ABA Treatment |
| GERI MARIA HARRIS (Texas Young Autism Project), Gerald E. Harris (Texas Young Autism Project) |
| Abstract: The 2008 Annual Report by the Texas Council on Autism emphasizes the need for widespread screening of young children for autistic spectrum disorders in order to identify those needing special services. One measure that has potential for cost-effective screening use is the Child Behavior Checklist (CBCL). The CBCL is a well established, valid, and reliable parent report measure of early childhood behavior. Previously presented research has demonstrated the very good inter-parental reliability, strong correlation with autistic spectrum disorders diagnostic criteria, and cognitive treatment outcome predictive utility of the CBCL. The present study looks at the CBCL as a predictor of post- ABA treatment undesirable characteristics and behaviors seen in ASD. Study participants were mothers of 215 young children diagnosed with an autistic spectrum disorder who received ABA treatment for one year. The CBCL and CARS (Childhood Autism Rating Scale) measures were administered, as part of an extensive evaluation battery, both before and after treatment. Statistical analyses, including cross-lagged panel correlation analysis, of the data showed that the CBCL can significantly predict decreases in the CARS scores post-treatment. Implications for identification and treatment of children with autistic spectrum disorders are discussed. |
| |
| Use of the WPPSI-III With Children With Autism: Revised Normalizations and Psychometric Properties for Interpretation |
| WENDY J. NEELY (Texas Young Autism Project), Gerald E. Harris (Texas Young Autism Project) |
| Abstract: Measurement of cognitive abilities of children with autism is integral to the design and evaluation of behavioral interventions. The Wechsler Preschool and Primary Scale of Intelligence is generally considered to be the “gold standard” of intelligence tests but the normalization tables were developed for general population preschool age children. Use of these tables for children with autism is problematic for assessing progress when cognitive gains are achieved at a slower rate than children in the general population. Test scores also frequently result in a “floor effect”. Previous research presented WPPSI-III norm tables developed specifically for the autism population which allow a more specific and accurate presentation and interpretation of a child’s change in cognitive functioning and a psychometric analysis of reliability across a one year time span. The present study extends the psychometric analysis for these norms through examination of differences across subtests for 220 repeated administrations. When used together, these new tools can provide useful information such as progress for an individual relative to a specific population and better identification of specific skill strengths and weaknesses. |
| |
| Reducing Variability: ABA Treatment Outcome Data for Children With Autism |
| GERALD E. HARRIS (Texas Young Autism Project), Wendy J. Neely (Texas Young Autism Project) |
| Abstract: A substantial body of research demonstrates that behavioral intervention (ABA) does improve the overall cognitive abilities of children with Autism. However, the goal is also to reduce the variability across intellectual skills that is such a prominent characteristic of autism. Significant differences across cognitive abilities can be a serious problem in itself, and interfere in successful participation in mainstream life. Previous research has not directly addressed this issue. The present study examines outcome data from a large sample of young children with autism who participated in a systematic behavioral treatment program for one year in order to determine the degree to which pretreatment cognitive variability was reduced. Pre-treatment and post-treatment cognitive test data from 220 young children was examined statistically to assess change over time in the variability of composite and subtest scores on the Wechsler Preschool and Primary Scales of Intelligence 3rd Ed. (WPPSI-III). Findings are consistent with previous research in that children with autism do exhibit increased cognitive ability following ABA intervention, and additionally show that, for many children, there is significant reduction in variability within composite scores and across subtests. The implications for diagnostic and clinical treatment outcome interpretations are discussed. |
|
| |