Subtest Scatter Research Articles

Measuring intelligence in Autism and ADHD: Measurement invariance of the-Binet 5th edition and impact of subtest scatter on abbreviated IQ accuracy.

Scatter and heterogeneity in cognitive profiles is thought to be common in autism spectrum disorder (ASD), which may indicate differences in the construct of IQ. However, less research has investigated IQ scatter in attention-deficit/hyperactivity disorder (ADHD). Scatter is also thought to negatively impact the predictive validity of IQ summary scores, although there is research refuting this notion. Abbreviated IQ tests, such as the Stanford-Binet fifth edition (SB-5) abbreviated battery IQ (ABIQ), may be especially susceptible to the influence of scatter. We tested the measurement invariance of the SB-5 as well as the predictive validity of the ABIQ in predicting FSIQ in 1679 youth (21% female) ages 2-16 years with a clinical diagnosis of ASD or ADHD. Results indicated the SB-5 is measuring IQ the same way in ASD and ADHD. There were no differences between diagnostic groups in scatter between ABIQ (i.e., routing) subtests. Additionally, scatter was not related to dimensional autistic traits. Higher degree of scatter was associated with poorer predictive validity of the ABIQ and a higher likelihood of overestimating FSIQ, regardless of diagnosis. Overall, we found more similarities than differences between the ASD and ADHD groups. Our results show that the SB-5 ABIQ is generally a strong predictor of FSIQ in youth with neurodevelopmental disorders. However, the use of the SB-5 ABIQ in research and clinical applications, without consideration of scatter on routing subtests, is potentially problematic.

Relative subtest scatter on the WAIS-IV in a clinical sample referred for outpatient neuropsychological evaluation

Objective: The primary aim of this study was to examine relative inter-subtest variability, or scatter, on the Wechsler Adult Intelligence Scale-Fourth Edition (WAIS-IV) in a clinical sample of patients referred for neuropsychological evaluation and compare the findings to corresponding data from the scale’s standardization sample. Method: Participants were 638 individuals diagnosed primarily with neurological, psychiatric, or neurodevelopmental disorders who completed the 10 core subtests of the WAIS-IV as part of a comprehensive assessment battery. Results: Mean participant scores on the WAIS-IV Full Scale IQ and all index composites were within the average range, overall, but were significantly below those of the standardization sample. The correlation between scatter range and highest subtest scaled score was significant, r = .65, indicating a greater degree of subtest scaled score variability in participants with higher than average peak subtest scaled scores than participants with average or below peak subtest scaled scores. Mean variability by highest subtest scaled score was, in most cases, larger in this clinical sample relative to the scale’s standardization sample. Exploratory secondary analyses also revealed specific differences in relative scatter based on diagnostic group classification. Conclusions: Subtest scatter on the WAIS-IV is common among both healthy individuals and clinical patients. Although somewhat higher in this investigation’s clinical sample, the significance of this finding generally appears to be of nominal value during interpretation of individual cases but may have some utility in formulating hypotheses when considered in conjunction with reliability data and other approaches for analyzing test scores. High scatter is not pathognomonic of abnormality, and at least some degree of caution is warranted when interpreting subtest scaled score differences on the WAIS-IV.

Normal Variability of Children’s Scaled Scores on Subtests of the Dutch Wechsler Preschool and Primary scale of Intelligence – Third Edition

Intelligence tests are included in millions of assessments of children and adults each year (Watkins, Glutting, & Lei, 2007a, Applied Neuropsychology, 14, 13). Clinicians often interpret large amounts of subtest scatter, or large differences between the highest and lowest scaled subtest scores, on an intelligence test battery as an index for abnormality or cognitive impairment. The purpose of the present study is to characterize “normal” patterns of variability among subtests of the Dutch Wechsler Preschool and Primary Scale of Intelligence – Third Edition (WPPSI-III-NL; Wechsler, 2010). Therefore, the frequencies of WPPSI-III-NL scaled subtest scatter were reported for 1039 healthy children aged 4:0–7:11 years. Results indicated that large differences between highest and lowest scaled subtest scores (or subtest scatter) were common in this sample. Furthermore, degree of subtest scatter was related to: (a) the magnitude of the highest scaled subtest score, i.e., more scatter was seen in children with the highest WPPSI-III-NL scaled subtest scores, (b) Full Scale IQ (FSIQ) scores, i.e., higher FSIQ scores were associated with an increase in subtest scatter, and (c) sex differences, with boys showing a tendency to display more scatter than girls. In conclusion, viewing subtest scatter as an index for abnormality in WPPSI-III-NL scores is an oversimplification as this fails to recognize disparate subtest heterogeneity that occurs within a population of healthy children aged 4:0–7:11 years.

Intelligence structure of 16–18 years old intellectually gifted students

The article analyses the intelligence structure of 16–18-year-old gifted students. Previous findings have indicated that high IQ individuals show more subtest scatter and a large discrepancy between abilities and very highly developed skills in some modalities, with many other cognitive skills being average or just slightly better. So, intelligence structure and subtests analysis, rather than IQ, became the focus of interpretation, and a multidimensional view of intelligence, provides a greater insight into the nature of human abilities than do summary intellectual indexes.Intelligence Structure Test 2000R (I-S-T 2000 R, Amthauer et al., 2001) was administrated in a group setting to 209 students aged 16–18 years from two schools in different Lithuanian towns. Data from 29 intellectually gifted students with Reasoning Total score 95 percentile and higher were included for analyses of intelligence structure.The exploratory factor analysis was applied to the I-S-T 2000 R basic module scores of intellectually gifted students to evaluate the validity I-S-T 2000 R for this population. A three-factor solution that mirrored the numerical, figural and verbal factors of the I-S-T 2000 normative sample appeared to be the most supportable.Intellectually gifted students had higher intellectual abilities, knowledge and memory skills than their peers. The mean of I-S-T 2000 R scale and subtest scores for this sample of gifted students was considerably above the normative mean. The significant differences in scores between gifted and norm groups were found for Calculations and Number Series subtests. Fluid (Gf) and Crystallized (Gc) Intelligence, Numerical (N) and Verbal (Vb) Intelligence discrepancies (Gf > Gc and N > Vb) for gifted students group were significantly greater discrepancies obtained in the standardization sample. Intellectually gifted students’ Verbal Memory scores wel significant higher Visual Memory scores and their Figural Knowledge scores was significantly higher than the Verbal Intelligence and Numerical Intelligence scores. The gifted students’ intellectual strengths are mathematical abilities, abilities to calculate, inductive and logical reasoning.

Relative Subtest Scatter in the WAIS-IV Standardization Sample

The frequencies of differences between highest and lowest subtest scores as a function of highest subtest score (relative scatter), are reported for the standardization sample of the Wechsler Adult Intelligence Scale-IV (WAIS-IV). Large differences between highest and lowest subtest scores were common. The degree of relative scatter was related to the height of the highest subtest score. For the 10 core WAIS-IV subtests, the correlation between the level of the highest subtest score and the amount of scatter was r = .62; for all 15 subtests the correlation was. 63. The level of the highest subtest score was more strongly related to scatter than was Full Scale IQ. Clinical implications for inferring cognitive impairment and estimating premorbid abilities are discussed. When considering the possibility of acquired cognitive impairment, we recommend caution in the interpretation of subtest score differences.

Intra-Individual Subtest Variability on the Dutch Wechsler Intelligence Scales for Children—Revised (WISC–RNL) for Children with Learning Disabilities, Psychiatric Disorders, and Epilepsy

It is common practice to look at disparities among subtest scores ("scatter") on an intelligence test to establish if a score is deviant. However, it remains unclear whether subtest scatter reflects primarily normal variation within individuals or is clinically meaningful. The present study explored this issue based on data from 467 children with developmental disabilities tested on the Dutch WISC-R(NL). Of these children, 132 had learning disabilities, 178 had psychiatric disorders, and 157 had epilepsy. Subtest scatter was defined as scaled-score range (highest minus lowest scaled score). When contrasted with "normal scatter," the overall sample revealed higher ranges on the Performance Scale and Full Scale, although effect sizes were small. Analysis of the data for the three separate clinical samples revealed unusual scatter only for the sample of children with psychiatric disorders. When comparing the clinical samples, scaled-score range was larger for the sample of children with psychiatric disorders than for those with epilepsy. Two distinct subsamples revealed elevated ranges with moderate effect sizes: children with autistic spectrum disorders and children with left hemisphere seizures. These results suggest that elevated subtest scaled-score range might characterize specific clinical samples rather than denoting an overall sign of pathology.

PSYCHOLOGICAL EVALUATION AND TREATMENT OF A CHILD WITH HYPERKINETIC DISORDER – CASE REPORT

Abstract Hyperkinetic disorders occur early in child development and are characterized by disturbed attention and hyperactivity displayed in various environments. Restlessness in dangerous situations and desinhibition in social settings with impulsive violation of social rules, are also commonly present. Although the disorder is conceptualized as being primarily biological, psychological treatment is also needed. Psycho diagnostic and anamnestic findings are used as a basis for treatment planning. With regard to the individual differences between children with hyperkinetic disorder, accurate assessment is crucial in order to meet the requirements for successful individualized treatment using different treatment options. Cooperation with parents and teachers is necessary. Key words: hyperkinetic disorder, psychological evaluation, psychotherapy, multimodal approach (ProQuest: ... denotes non-US-ASCII text omitted.) Introduction Hyperkinetic disorders are a part of a group of developmental disorders whose key features are disturbed attention and hyperactivity which are evident in at least two different settings (for example, school and home). Attention disturbances are manifested in discontinuation of tasks and leaving projects unfinished. Hyperactivity involves excessive restlessness, especially in situations which call for serenity, as well as engaging in activities which are disorganized and serve no purpose. Lack of inhibition in social settings with impulsive violation of social rules is also common (1), as well as emotional instability (low frustration tolerance, frequent mood swings) (2). Hyperkinetic disorders always occur early in development, but are rarely recognized before the age of three. Starting school, these problems become more evident since the school setting calls for a relative calmness and a higher level of self-control, which they lack. Although the disorder is conceptualized as primarily biological (3), drug treatment alone did not prove itself useful (4). The disorder needs to be tackled psychotherapeutically as well, especially with regard to its secondary consequences like low academic achievement, peer and family problems, antisocial behavior (5). Case report A boy aged 8 years and 2 months, a second grade student, was referred to specialist assessment by the school psychologist. Referral grounds were stated as observed difficulties when compared to his age group, evident even in pre-school and manifested in aggression and hyperactivity of the child. The boy lives with his mother. Parents were divorced three years ago, father is an alcoholic. Pregnancy and child birth were normal, as was the child's early psychomotor development, according to his mother. The boy was restless in pre-school, disrupted group activities and other children avoided playing with him. Problems became exacerbated as he started school. He is restless in class, leaves his desk, fidgets, interferes with activities of other children, talks a lot, laughs and makes other children laugh. Children often complain to their teacher about his behavior and exclude him from group activities. He manages to fulfill most school requirements with a relative success. At first encounter with the boy, verbal contact is established. He shows high motivation when presented with test materials and tries to accommodate examiners requirements. But after approximately 15 minutes he becomes impatient, hasty, fidgets, has a hard time keeping his attention focused on presented material. His achievement on Intellectual functions test (Revisk) is average (IQ=96) (6). There is a remarkable dominance of nonverbal, manipulative competencies (MIQ=103) compared to verbal (VIQ=89), which combined with problems in focusing his attention on tasks, points to his educative stimulation being lacking or inadequate. Subtest scatter is mildly uneven with a significant raise in short-term memory, and decline in understanding of social situations. …

Wechsler Adult Intelligence Scale-III Inter-Subtest Scatter: A comparison of brain-damaged patients and normal controls

This study assessed the validity of Wechsler Adult Intelligence Scale-Third Edition (WAIS-III) inter-subtest scatter (as measured by the scaled score range) as an indication of cognitive impairment due to brain dysfunction. Participants were 174 individuals with documented brain damage. Means for age, education, and WAIS-III full-scale IQ were 49.19 years (SD = 15.33), 12.57 years, (SD = 2.78), and 88.45 (SD = 17.78). The scatter ranges for brain-damaged participants were compared with those of the WAIS-III standardization sample. Results indicated that the full-scale IQ is significantly correlated with amount of scatter, r(172) = .42, p < .0001, so that a definitive answer to the question requires control of full-scale IQ. In the present analyses, inter-subtest scatter among the individuals for the sample as a whole was no greater than that for persons in the standardization sample, but subtest scatter was significantly greater for participants with IQs > or = 90. This finding suggests that there may be small differences in scatter between brain-damaged persons and normal controls, but that scatter is relatively insensitive to the presence of brain damage or dysfunction. Therefore, interpretation of marked inter-subtest variability as a sign of brain damage appears unwarranted.

DISCRIMINATION OF HEALTHY AND BRAIN-DAMAGED ELDERLY USING THE WAIS-R

The authors evaluated the ability of WAIS-R variables to discriminate normal from brain-damaged persons 75 years of age and older. Fifty-four brain-damaged patients and 54 healthy elderly were matched on the variables of age, education, sex, and race. Significant level of performance differences emerged between the groups on all subtests and the three IQs. No group differences were found on the VIQ-PIQ discrepancy score or range of subtest scatter. Step-wise discriminant analysis revealed that Digit Symbol was the most potent WAIS-R variable, accounting for 34% of the variance and correctly classifying 79.6% of the subjects.

The dissociation of perception and cognition in children with early brain damage

Reduced non-verbal compared to verbal intelligence is used in many outcome studies of perinatal complications as an indication of visual perceptual impairment. To investigate whether this is justified, we re-examined data sets from two previous studies, both of which used the visual perceptual battery L94. The first study comprised 47 children at risk for cerebral visual impairment due to prematurity or birth asphyxia, who had been administered the McCarthy Scales of Children's abilities. The second study evaluated visual perceptual abilities in 82 children with a physical disability. These children's intellectual ability had been assessed with the Wechsler Intelligence Scale for Children-Revised and/or Wechsler Pre-school and Primary Scale of Intelligence-Revised. No significant association was found between visual perceptual impairment and (1) reduced non-verbal to verbal intelligence; (2) increased non-verbal subtest scatter; or (3) non-verbal subtest profile deviation, for any of the intelligence scales. This result suggests that non-verbal intelligence subtests assess a complex of cognitive skills that are distinct from visual perceptual abilities, and that this assessment is not hampered by deficits in perceptual abilities as manifested in these children.

Incremental validity of WISC–III profile elevation, scatter, and shape information for predicting reading and math achievement.

The use of cognitive subtest profiles to hypothesize about children's learning strengths and weaknesses implicitly assumes that subtest profiles are predictive of academic performance. To test this assumption, Wechsler Intelligence Scale for Children-Third Edition (WISC-III) subtest profiles were decomposed into elevation, scatter, and shape components and sequentially regressed onto reading and math achievement scores for 1,118 nonexceptional and 538 exceptional students. Profile elevation was statistically and practically significant for both exceptional (R = .36-.61) and nonexceptional (R = .72-.75) students. Profile scatter did not aid in the prediction of achievement. Profile shape accounted for an additional 5%-8% of the variance in achievement measures. It was concluded that using WISC-III subtest scatter and shape to predict academic performance was not supported by the accumulated scientific evidence.

Relationship between psychopathology, learning disabilities, or both and WISC-III subtest scatter in adolescents

The aim of this study was to determine whether subtest scatter might be an indicator of learning disabilities (LDs), psychopathology, or of both, in a sample of normal adolescents. Subjects completed the WISC-III and the MMPI-A and then their scores for the WISC-III subtests, verbal scatter (VScatter), performance scatter (PScatter), full scale scatter (FScatter) scores, perceptual organisation factor (PO), and the SCAD profiles were calculated. The MMPI-A was scored and the clinical scales were extracted. The hypothesis that subtest scatter is associated with LDs was generally not supported. There was a relationship between subtest scatter and psychological disturbance in adolescent males, however the results for females were less clear. Further, high PScatter in adolescent males was associated with an MMPI-A codetype 4-2/2-4, indicating tendencies toward depressive features, delinquent behavior, and possibly substance abuse, tendencies which are likely to lead to school failure. © 1999 John Wiley & Sons, Inc.

WAIS-R performance following closed-head injury: A comparison of the clinical utility of summary IQs, factor scores, and subtest scatter indices

WAIS-R performance can be analyzed at the level of the IQs, factor scores, or subtests. This study examined the relative utility of such analyses in a closed-head-injured (CHI) sample (n = 233). Comparison of the sample with matched controls (n = 117) indicated that factor scores may be the most appropriate level of analysis in clinical practice. This conclusion was supported by a series of discriminant function analyses performed on the WAIS-R scores of the CHI sample and 356 healthy controls drawn from the general population. Factor scores achieved a significantly higher classification accuracy than IQs or indices of subtest scatter (all ps < .001). Combining the scatter indices with the composite measures (i.e., 1Qs or factor scores) did not produce a significant improvement in discrimination over that achieved by the composites alone. Counter to the assumptions underlying the clinical analysis of subtest profiles, CHI participants failed to exhibit significantly greater intraindividual subtest variability in their subtest profiles than the matched control sample.

Methods of Estimating Premorbid Functioning

Clinical neuropsychological assessment frequently requires the comparison of obtained scores against some estimate of premorbid level of functioning, but only recently has significant attention been turned to objective methods to accomplish this objective. Clinical judgment, although useful in some circumstances, is generally insufficient. Other methods of estimating premorbid function include demographic regression formulae, such as the Barona formula, subtest scatter methods, such as that suggested by Lezak, and the use of current scores on tests of presumably spared abilities, such as the National Adult Reading Test (NART). Almost all methods predict to some general level of intellectual functioning rather than to specific neuropsychological skills. This paper reviews the suggested methods in terms of the underlying assumptions and the available empirical evidence. Suggestions for future research include the development of skill specific predictors as well as investigations regarding the relation between predictor accuracy and characteristics of the subject, such as high versus low premorbid functioning in the subject. Additionally, there is a great need for methods to predict premorbid functioning in children. © 1997 National Academy of Neuropsychology. Published by Elsevier Science Ltd

Base rate data on the abnormality of subtest scatter for WAIS-R short-forms.

Tables are presented which permit clinicians to determine the abnormality of a WAIS-R subtest profile when one of nine different short-forms has been administered. The tables record the discrepancy, between each subtest and the individual's mean subtest score, which must be exceeded to occur in less than 15 per cent or 5 per cent of the healthy population. Two tables are provided; the first uses a statistical formula developed by Silverstein (1984a) to estimate the discrepancies from the US standardization sample. The second presents empirically derived discrepancies based on a sample of 326 healthy individuals which was representative of the adult UK population in terms of age, sex and social class distributions. The UK sample was also used to generate base rate data on the cumulative number of abnormal subtest deviations exhibited by healthy individuals.

WISC-III subtest scatter as a function of highest subtest scaled score.

WISC-III subtest scatter as a function of highest subtest scaled score.

WISC-III subtest scatter as a function of highest subtest scaled score.

The Wechsler Intelligence Scale for Children-III (WISC-III) standardization data were analyzed to determine the frequency of occurrence for relative intersubtest scatter ranges. The relative scatter range was defined as the difference between the highest and lowest subtest scaled scores based on the value of the highest subtest scaled score. This approach to intersubtest scatter provides the greatest precision in scatter analysis to help clinicians determine whether the amount of scatter obtained in WISC-III clinical profiles is rare enough to be considered abnormal. For ease of clinical use, tables are provided that report frequency of occurrence for the lowest subtest scaled score by each level of highest subtest scaled score. Potential scatter range, rather than overall level of intelligence, was the primary determinant of the magnitude of scatter; the greater the possible range, the larger the magnitude of scatter.

WAIS-R subtest scatter: base-rate data from a healthy UK sample.

Base-rate data on subtest scatter for the WAIS-R was obtained from a sample of 200 healthy subjects recruited to match the adult UK population in terms of age, sex and social class distribution. Tables are presented which permit an assessment of the abnormality (i.e. rarity) of an individual's pattern of WAIS-R performance in terms of subtest range and subtest deviations from an individual's subtest mean. Guidance on the appropriate use of the tables is offered and the data are compared with data from the US WAIS-R standardization sample where appropriate. The distinction between reliable and abnormal differences is highlighted.

Testing: Exploring the WISC‐III as a measure of giftedness∗

This study investigated patterns of WISC‐III IQ subtest scores for gifted children. Subtest scatter of WISC‐III scores occurred with greater frequency in a gifted sample than for subjects reported in the test manual's normative sample. Variability from the subtest mean of the gifted was most evident on Similarities, Comprehension, Coding, and Symbol Search. The extent of subtest scatter and its pattern on the WISC‐III was similar to typical patterns of highs and lows found for gifted children on the WISC‐R with one notable exception. There was a marked lack of strength in Block Design, previously seen as a peak subtest for gifted students when assessed with the WISC‐R.

Base rates for the Selz-Reitan. Index of subtest scatter on the WPPSI-R

Using the standardization sample of the Wechsler Preschool and Primary Scale of Intelligence-Revised (WPPSI-R) an analysis was done to determine the base rates of occurrence of the Selz-Reitan Index (SRI) of unusual or abnormal scatter. Findings offered support for the use of the WPPSI-R to determine unusually or abnormally large SRIs in clinical practice. However, It was found that the clinical interpretation of obtained indexes depend upon the Full Scale IQ of the individual and the level of the SRI selected for interpretation. Further research on clinical neurological populations is encouraged.