Secondary Logo

Predictors and Correlates of Adaptive Functioning in Children with Developmental Disorders

Stephens, Tracy J. PhD

Departments: Critical Reviews of Current Research

University of South Dakota

Critical Reviews of Current Research: Manuscripts for this department should be sent directly to Ann F. VanSant, PhD, PT, Temple University, Department of Physical Therapy, College of Allied Health Professions, 3307 N. Broad Street, Philadelphia, PA 19140.

Predictors and Correlates of Adaptive Functioning in Children with Developmental Disorders

by M. Liss, D. Fein, D. Allen, et al. Journal of Autism and Developmental Disorders 2001;31: 219–230.

Back to Top | Article Outline


Autism, recently suggested to be a genetic disorder, occurs in one or two of every 1000 births. Frequently autism results in impairments that affect adaptive functioning. The authors of this study looked at the important relationships between cognitive functioning, autistic symptomatology, language, memory, and adaptive behavior in four groups of children with developmental disorders. In this complicated study the authors examined five hypotheses: 1) Children with autism have more significant impairments in adaptive functioning than children without autism but with similar cognitive abilities; 2) IQ is related to the development of adaptive functioning, especially in children who are more significantly impaired; 3) autism symptomatology would affect adaptive functioning primarily in the areas of socialization and communication; 4) communication skills are highly correlated with performance on language tests; and 5) memory plays an important role in gaining daily living skills.

Back to Top | Article Outline


The participants in this study were children being followed by the authors as part of a large longitudinal project. At preschool age, the participating children were referred from five sites in four different states. The children were classified into four groups following a comprehensive developmental evaluation that included a screening instrument, direct observation, parent and teacher interviews, performance on language assessments, and diagnoses by experienced child psychiatrists. The two study groups included 1) children with autism and nonverbal IQ scores of 80 or below and 2) children with autism with nonverbal IQ scores above 80. The two contrast groups included 1) children without significant autism symptomatology with nonverbal IQ scores of 80 or below and 2) children without significant autism symptomatology with nonverbal IQ scores above 80 and language impairment, defined as a score less than 80 or at least one standard deviation below the mean on the Test of Early Language Development (see Table 1 in the article).

For this study, the children were reevaluated at age nine years. It was noted that the diagnosis of autism was retained from preschool for all of the children; however, at the time of reevaluation, the performance on nonverbal IQ measures were used to categorize the children into high- or low-functioning groups. This yielded 123 children (75 with autism and 48 without autism).

For this study, the participants were administered a battery of assessments, which varied depending on their level of ability. The following instruments were available for use: Vineland Adaptive Behavior Scales, Stanford Binet (fourth edition), Wing Autism Diagnostic Interview Checklist, the Peabody Picture Vocabulary Test–Revised, the Expressive One-Word-Picture-Vocabulary Test, California Verbal Learning Test, the Test of Written Spelling, the Word Identification Subtest of the Woodcock Reading Mastery Test, and the Token Test. Both descriptive statistics and sequential regression analysis (SPSS software) were used to find correlations within the data. Listwise deletion (removal of the subject’s data from the analysis if any information for that person was missing) was used in calculating the regressions for this study. Composite scores for the Vineland and the Stanford Binet tests were calculated but were not used for the analyses in this study.

Back to Top | Article Outline


The following findings were noted to relate to the five hypotheses that the authors set out to assess:

  1. Children with autism have more deficits in adaptive functioning in the areas of socialization and daily living than do other children with similar IQ scores; this was evident in both groups of children with autism. Communication was found to not be more impaired in children with autism than in the similarly matched children without autism.
  2. The correlation between adaptive functioning and IQ was strongest for children with and without autism who have lower IQ scores. In children with autism who are high functioning, the adaptive behavior correlates indicate that there is more of a difference between the IQ score and the adaptive behavior scores, especially in the areas of socialization and daily living skills, compared with children with similar IQ scores and developmental language disorder.
  3. The degree of autism symptoms seemed to relate to adaptive behavior of only individuals with autism who were high functioning. In individuals with autism and lower nonverbal IQ scores, the degree of autism seemed to be related only to the social domain and autistic social behavior.
  4. The hypothesis that communication skills are highly correlated with performance on language tests was supported for all groups studied.
  5. Memory was found to possibly play an important role in the learning of daily living skills and other adaptive domains for children with autism who have nonverbal IQ scores above 80. This correlation was not found in the developmental language delay contrast group.
Back to Top | Article Outline


The authors of this study raised several interesting discussion points. One of these includes the possibility of using adaptive behavior measures as estimates of IQ in children with autism who are lower functioning. As the authors point out, this has been suggested in the literature before and is confirmed by the results of the current study. The authors also point out that there may be a difference in how adaptive behavior is used to predict functioning level and learning in these two very different groups of individuals with autism. In individuals who are lower functioning, nonverbal IQ may be a good predictor of their overall adaptive behavior functioning and may actually be a rate-limiting factor for the development of some of these skills. For individuals functioning at this lower level, both IQ and adaptive behavior possibly measure the same skill area (ability to understand and learn simple tasks). For individuals who are higher functioning, language, memory, specific learning strategies, and autism symptoms or other negative behaviors may play a much larger part in affecting the development of adaptive behavior skills.

Back to Top | Article Outline

Limitations and Implications

The authors stated that the conclusions drawn from this study are additional hypotheses that merit further research and should not be viewed as definitive statements. The lack of complete data for numerous participants noted in the results may be problematic as a whole, especially for individuals in the two contrast groups, because the data of those participants would have been deleted during the regression analysis. The individuals with autism who were low functioning and the individuals with low IQ in the contrast group were matched only on the nonverbal IQ scores because verbal IQ measures could not be obtained on 17 of the individuals with autism; thus, the verbal IQ score reported is an overestimate for the actual sample. The individuals participating in this study were not matched for types of intervention programs. It is possible that the group and individual differences reported in this sample may have been affected by different interventions. Likewise, it is noted that some children were categorized into different groups at age nine years than they were as preschoolers at the time of entry into the longitudinal study. The vast developmental changes of those children who may have been reclassified from lower functioning to high functioning may be suggestive of additional variables that may have affected group and individual differences.

Overall, this study provides invaluable information on a large number of variables influencing how adaptive behavior is related to other aspects of learning and development in different groups of children with developmental disorders. Adaptive behavior is a very important indicator of the level of independent functioning for all children with disabilities. As the authors of this study so rightly point out, it is the level of skill development in this area and not cognitive skill that determines the prognosis for the majority of individuals with autism. It is critical that methods be developed that allow us to better measure and predict adaptive functioning in this population. Because children with autism, both higher and lower functioning, have significant deficits in adaptive behavior, this is a key area in which service providers in all professions should focus their interventions. Understanding the importance of adaptive behavior skills and being able to identify specific areas of strengths and weakness as well as limiting factors (such as nonverbal IQ in individuals who are lower functioning) may prove to be very helpful in developing good educational goals and objectives. This is especially true for professionals in the related service areas of occupational and physical therapy.

Tracy J. Stephens, PhD, University of South Dakota

© 2001 Lippincott Williams & Wilkins, Inc.