Early Childhood Measurement and Evaluation
Tool Review
Early Childhood Measurement and Evaluation (ECME), a portfolio within CUP, produces
Early Childhood Measurement Tool Reviews as a resource for those who conduct screening,
assessment, and evaluation. To learn more about ECME and CUP, provide feedback, or to
access additional reviews, visit our website at [Link] or email us at
cup@[Link]
Adaptive Behaviour Assessment System System -
Second Edition (ABAS-II)
Measurement Areas:
The Adaptive Behavior Assessment System ‐ Second Edition (ABAS‐II) is a norm referenced tool
designed to assess adaptive skills in individuals from birth to 89 years of age.
The tool measures 10 skill areas:
1) Communication
2) Community Use
3) Functional Academics
4) Home Living
5) Health and Safety
6) Leisure
7) Self‐Care
8) Self‐Direction
9) Social
10)Work (Optional)
Purpose:
The ABAS‐II can be used to assess an individual’s adaptive skills:
to assist in the diagnosis and classification of disabilities and medical/clinical disorders;
for the identification of adaptive skill strengths and difficulties in a person’s daily living
environment;
for the identification of service needs in treatment or intervention programs;
for research related to adaptive skill progression (i.e., intervention program evaluations).
Length and Structure:
The ABAS‐II consists of a series of 5 record forms: Parent/Primary Caregiver (for children ages 0‐
5), Teacher/Daycare Provider (for children ages 2‐5), Parent (for children ages 5‐21), Teacher (for
children ages 5‐21), and Adult (ages 16‐89). The examiner gives selected forms to respondents
such as parents, primary caregivers, teachers, daycare workers, counselors, who are familiar with
the daily living circumstances of the individual. Each of the respondents fills out a questionnaire
that assesses their knowledge of the person’s adaptive skills. Each item can be answered “0” (is
A01 1 of 5
not able to perform the task), “1” (never or almost never performs the task), “2” (performs the
task sometimes), or “3” (always or almost always performs the task); the respondent also
indicates for each item whether her/his response was a guess or an estimate.
Raw scores are calculated by tabulating totals from the items answered in each skill area. The raw
scores can be converted into composite scores, a skill profile, scaled scores, percentile ranks,
domain‐specific scores, descriptive classifications, and test‐age equivalents. According to the
manual, each respondent form in the ABAS‐II should take between 20‐30 minutes to complete.
The examiner is responsible for scoring the ABAS‐II not the responder and this takes
approximately 10 minutes per form.
Materials:
The publisher classifies the ABAS‐II as a “B‐level” qualification. The publisher requires the
purchaser to fit into one of the following categories: (a) trained and certified by a recognized
institution in a relevant area of assessment (with or without a Master’s degree), (b) a member of a
Speech‐Language‐Hearing Association or an Occupational Therapy Association, or (c) possess a
Master’s (or Doctorate) degree in psychology, education, or relevant field with training in
assessment.
The ABAS‐II is available in a variety of packages from the publisher. The complete Examination Kit
consists of the manual, 5 parent/primary caregiver forms (for ages 0‐5), 5 parent forms (for ages
5‐21), and 5 adult forms (ages 16‐89), for CDN $280. The publisher also supplies kits targeted to
specific groups, such as teachers, adults, and infants/preschool children for CDN $290; each of
these kits contain additional scoring forms. Extra scoring forms are also available through the
publisher in packages of 25 or 100. The ABAS‐II Scoring Assistant software (Windows‐only) is
available for CDN $225.
Accessibility:
The complete ABAS‐II is available in the English language; Spanish respondent forms are also
available. While the ABAS‐II is most frequently used to measure adaptive skills in individuals with
cognitive and developmental delays, the manual notes that the tool is useful for individuals
diagnosed with many other kinds of disabilities, clinical disorders, and medical or neurological
disorders.
Administration, Scoring, and Interpretation:
According to the manual, the tool can be administered and scored by professionals or para‐
professionals who have received extensive training in ABAS‐II administration and scoring. A
professional with formal coursework in educational assessment must interpret the scores. The
test is easy to administer, easy to score, and moderately difficult to interpret based on the
interpretation guide in the manual.
Subscales:
A01 2 of 5
The ABAS‐II consists of 9 or 10 skill areas (see Measurement Areas section for list) that are
combined to create the “General Adaptive Composite” (GAC) score that represents a norm
referenced score for the individual. Each of the skill areas are also grouped into 3 adaptive
domains: Conceptual (Communication + Functional Academics + Self‐Direction), Social
(Leisure + Social), and Practical (Community Use + Home Living + Health and Safety + Self‐Care).
The “Work” skill area is considered to be a separate, optional domain. Since several skill areas do
not apply to certain age groups (i.e. Work and Home Living Skills are not relevant to a toddler), the
GAC score is calculated based on only applicable scales.
Documentation:
The manual included in the ABAS‐II kit contains specific procedures for administration, scoring,
and a very comprehensive section on interpretation. The manual also includes extensive sections
describing test standardization, norm development, and many kinds of reliability and validity.
Norming Sample:
The ABAS‐II was normed and standardized based on a representative sample of the
English‐speaking U.S. population ages 0 – 89 years of age using 1999‐2000 census data.
Infant‐preschool norms consisted of 750 teacher/daycare provider and 1,350
parent/primary caregiver completed forms.
School‐age norms consisted of 1,690 teacher and 1,670 parent completed forms.
Adult norms consisted of 990 self report and 920 rated by others completed forms.
The sample was stratified by sex, race/ethnicity, geographic region, clinical cases, and education
level. The examiner’s manual discusses the demographic characteristics of the sample in detail,
including information on the respondent sample. Since the ABAS‐II is the newest version of the
Adaptive Behavior Assessment System, academic reviews or studies of the test are not yet
available.
Reliability:
The ABAS‐II manual discusses four kinds of reliability measures:
Internal Consistency: Reliability scores ranged from .97 to .99 for the GAC (General Adaptive
Composite) scores, .91 to .98 for the adaptive domains (i.e., Conceptual, Social), and .80 to .97 for
the 10 individual skill areas. The authors note that the correlation coefficients were also high for
individuals with clinical diagnoses and all age groups, indicating that the test shows a high degree
of consistency for examinees with a wide range of adaptive functioning.
Testretest Reliability: A sub‐sample of examinees (569 children; 118 adults) was retested with
the same form of the ABAS‐II two days to six weeks after the first administration. Correlations
between both test scores were calculated, and yielded reliability coefficients in the .90s (excellent)
for GAC scores, .80s to .90s for adaptive scores, and .70s to .90s for the skill areas.
Interrater Reliability: The authors performed an analysis of inter‐rater consistency on a
stratified sub‐sample of examinees (257 children; 52 adults) and their respondents (2
teachers/daycare providers, 2 parents/primary caregivers, 2 teachers, and 2 adults). Interrater
A01 3 of 5
reliabilities varied across each of the forms used. In general, inter‐rater reliabilities across all
respondent forms were good, between .82 and .91 for the GAC, .78 and .84 for adaptive domains,
and .70 to .82 for the skill areas.
CrossForm Reliability: The authors performed correlative analyses to determine the
consistency of ratings between teachers/daycare providers and parents/primary caregivers (of
the same child), and also between adults as rated by others and adults who rated themselves.
According to the authors, correlations between teacher/daycare provider and parent/primary
caregiver scores averaged between .68 and .70 for the GAC, .62 and .70 for adaptive domain scores,
and .51 and .66 for skill areas – all moderate to good reliabilities. The correlations between self‐
reporting adults and adults‐as‐reported‐by‐others averaged at .95 for the GAC, .91 for adaptive
domain scores, and .88 for skill areas.
Validity:
The ABAS‐II manual has an extensive discussion of test validity. According to the authors, the
ABAS‐II is based on a theoretical foundation derived from the American Association of Intellectual
and Developmental Disabilities that includes the 10 skill areas and domain groupings discussed
earlier. The manual contains a description of the specific item‐ and content‐selection processes, as
well as detailed discussions of several validation studies.
Age Group Differences: According to the authors, the ABAS‐II shows strong sensitivity to age
group; thus demonstrating that the tool matches the typical adaptive skill acquisition process that
we would expect in an individual’s developmental lifespan.
Construct and Convergent Validity: According to the authors, the ABAS‐II was founded upon the
theoretical basis that each of its skill areas should be minimally related to each other, highly
related to their respective adaptive domains, and strongly correlated with the GAC. Across all
forms the intercorrelations between skill areas were .40s to .70s, the averaged intercorrelations
between skill areas and adaptive domains were .60s to .70s, and the intercorrelations between
skill areas and the GAC were .60s to .80s. This profile of results suggests that the ABAS‐II fits the
theoretical basis that it was designed with.
Factor Structure: The authors performed a factor analysis on the entire standardization sample
data set, yielding a single‐factor structure. According to the authors this would support the notion
that the ABAS‐II assesses a single factor related to general adaptive skill.
Concurrent Validity: The ABAS‐II manual contains specific validation studies that were
performed by the authors, demonstrating the concurrency between the tool and other related
scales. For the adaptive behavior‐related scales, small samples (< 60 children in each sample)
were used to compare scores on the ABAS‐II with the Vineland Adaptive Behavior Scale (VABS‐CE),
VABS Interview Edition (VABS‐IE), Scales of Independent Behavior‐Revised (SIB‐R), and the
Behavior Assessment Scale for Children (BASC).
The correlation between the General Adaptive Composite (GAC) on the ABAS‐II and the
VABS‐CE “Adaptive Behavior Composite” was .75 for the Teacher/Daycare Provider form,
and .84 for the Teacher Form.
The correlation between the GAC and the VABS‐IE Adaptive Behavior composite was .70.
A01 4 of 5
The ABAS‐II GAC was correlated (.57) with the SIB‐R “Broad Independence standard
score.”
The ABAS‐II was compared to the BASC – a tool designed to measure clinical dimensions of
behavior such as emotional and behavior disorders in children. The correlation between
the GAC and the BASC “Adaptive Skills Composite” was .80.
ABAS‐II scores were also compared to intelligence and achievement scale scores on the
WPPSI‐III, WISC‐III, WISC‐IV, WAIS‐III and WASI. While these validation studies are not
discussed in detail in this review, measures of intelligence were generally correlated in
the .40s and .50s, and measures of achievement were generally correlated in the .60s.
Sensitivity and Specificity: Clinical populations of adults and children were tested to determine
if the ABAS‐II could correctly discriminate between positive and negative clinical cases.
In general, children with clinical diagnoses (e.g., developmental delays, motor impairments,
receptive/expressive language deficits) typically showed performance at least 2 standard
deviations below the mean of a control group.
Similarly, adults with diagnoses (e.g., neuropsychological disorders, Alzheimer’s disease)
typically performed at least 2 standard deviations below the mean compared to control
groups.
According to the authors, this would suggest that the ABAS‐II demonstrates good
sensitivity for clinical populations, as it can correctly distinguish people diagnosed with
clinical disorders from a control group that was not diagnosed with clinical disorders.
According to the authors, the ABAS‐II also demonstrates good specificity, as it was able to
discriminate between differing levels of disability in children and adults. For example,
children and adults who were diagnosed with a mild intellectual disability often scored 2‐3
standard deviations higher in several skill areas, than children and adults who were
diagnosed with a moderate intellectual disability.
It should be noted that since the ABAS‐II is a newly revised test, no outside academic reviews or
analyses of this scale are available and this review is based on information provided by the
authors in the manual.
Publication Information:
The Adaptive Behavior Assessment Scale ‐ Second Edition, was developed by Patti L. Harrison and
Thomas Oakland. This review is based on the 2nd edition published in 2003 by Harcourt
Assessment.
References:
Publisher’s website: [Link]
Harrison, P.L., & Oakland, T. (2003). Adaptive Behavior Assessment – Second Edition Manual (ABAS
II). San Antonio, TX: Harcourt Assessment.
__________________________
How to cite this document:
This document was created by CUP and to cite this document use the following:
Community‐University Partnership for the Study of Children, Youth, and Families (2011). Review
of the Adaptive Behaviour Assessment System. Second Edition (ABASII). Edmonton, Alberta, Canada.
A01 5 of 5
