STRUCTURE OF THE DAS

The DAS-II consists of a cognitive battery of 20 subtests, covering an age range of 2 years, 6 months through 17 years, 11 months (2:6 through 17:11). The battery is divided into two overlapping age levels: (1) The Early Years battery is normed from age 2:6 through 8:11, with a usual age range of 2:6 through 6:11; (2) The School-Age battery is normed from age 5:0 through 17:11, and has a usual age range of 7:0 through 17:11. With those overlaps between the Early Years and the School Age batteries, it will be seen that the DAS-II Early Years and School-Age batteries were conormed for children ages 5:0 through 8:11 and therefore have a four-year normative overlap. (See Rapid Reference 1.2 for a description of the DAS-II subtests.)

The Early Years battery is further divided into two levels, lower and upper. The Lower Early Years level is most appropriate for young children ages 2:6 through 3:5, although it may also be used with older children with special needs. The Upper Early Years level is suitable for children normally in the age range of 3:6-6:11, although it may also be used with children up to age 8:11 if they have difficulty with the materials in the School-Age battery.

The DAS-II battery yields a composite score called General Conceptual

Rapid Reference I.I

DAS-II Batteries

Author: Colin Elliott

Publication date: 2007

What the test measures: Verbal (Gc), Nonverbal Reasoning (Gf ), Spatial (Gv), Working Memory (Gsm), Processing Speed (Gs), Phonological Processing (Ga), Recall of Objects (Glr), and General Conceptual Ability (GCA), which is a measure of the general factor g.

Age range: 2:6-17:11

Average Administration time: Six core subtests to obtain three clusters and GCA score = 31-40 minutes. Diagnostic subtests—School Readiness = 17 minutes, Working Memory = 12 minutes, Processing Speed = 9 minutes, Phonological Processing = 10 minutes.

Qualification of examiners: Graduate- or professional-level training in psychological assessment

Computer program: Scoring program included as well as a CD, which includes help in administering the Phonological Processing subtest and also useful demonstrations of administering the test using American Sign Language.

Publisher: The Psychological Corporation

A division of Pearson
555 Academic Court
San Antonio, TX 78204-2498
Ordering phone number: 800-211-8378
http://www.psychcorp.com
Web site: www.DAS-II.com

Rapid Reference 1.2

DAS-II Subtests

Verbal Subtests

• Verbal Comprehension: following oral instructions to point to or move pictures and toys.

• Naming Vocabulary: naming pictures.

• Word Definitions: explaining the meaning of each word. Words are spoken by the evaluator.

• Verbal Similarities: explaining how three things or concepts go together, what they all are (e.g., house, tent, igloo; love, hate, fear)

Nonverbal Reasoning Subtests

• Picture Similarities: multiple-choice matching of pictures on the basis of relationships, both concrete (e.g., two round things among other shapes) and abstract (e.g., map with globe from among other round things). [Nonverbal Cluster in Lower Early Years battery]

• Matrices: solving visual puzzles by choosing the correct picture or design to complete a logical pattern.

• Sequential and Quantitative Reasoning: figuring out sequential patterns in pictures or geometric figures, or common rules in numerical relationships.

Spatial Subtests

• Copying: drawing pencil copies of abstract, geometric designs.

• Recall of Designs: drawing pencil copies of abstract, geometric designs from memory after a five-second view of each design.

• Pattern Construction: imitating constructions made by the examiner with wooden blocks, copying geometric designs with colored tiles or patterned cubes. There are time limits and bonus points for fast work. An alternative, “untimed” procedure uses time limits but no speed bonuses. [Nonverbal Cluster in Lower Early Years battery]

Diagnostic Subtests

• Early Number Concepts: oral math questions with illustrations—counting, number concepts, and simple arithmetic.

• Matching Letter-Like Forms: multiple-choice matching of shapes that are similar to letters.

• Recall of Digits Forward: repeating increasingly long series of digits dictated at two digits per second.

• Recall of Digits Backward: repeating, in reverse order, increasingly long series of digits dictated at two digits per second.

• Recognition of Pictures: seeing one, two, or three pictures for five seconds or four pictures for ten seconds and then trying to find those pictures within a group of four to seven similar pictures.

• Recall of Objects—Immediate: viewing a page of 20 pictures, hearing them named by the evaluator, trying to name the pictures from memory, seeing them again, trying again to name all the pictures, and repeating the process once more. The score is the total of all the pictures recalled on each of the three trials, including pictures recalled two or three times.

• Recall of Objects—Delayed: trying to recall the pictures again on a surprise retest 15 to 20 minutes later.

• Speed of Information Processing: the student scans rows of figures or numbers and marks the figure with the most parts or the greatest number in each row. The score is based on speed. Accuracy does not count unless it is very poor.

• Phonological Processing: rhyming, blending sounds, deleting sounds, and identifying the individual sounds in words.

• Rapid Naming: naming colors or pictures as quickly as possible without making mistakes. The score is based on speed and accuracy

• Recall of Sequential Order: sequencing, from highest to lowest, increasingly long series of words that include body parts, and for more difficult items, non-body par ts.

Ability (GCA), which provides an estimate of overall reasoning and conceptual abilities. In addition, for ages 3:6 to 17:11, a Special Nonverbal Composite (SNC) is available and derived from the nonverbal core subtests appropriate for each battery level. The DAS-II also provides lower-level composite scores called cluster scores that are derived from highly g-saturated core subtests. Finally, there are numerous diagnostic subtests and clusters that measure other specific abilities. These diagnostic subtests do not contribute to the GCA or SNC, but give additional information about cognitive strengths and weaknesses. The overall structure is summarized in Figure 1.1.

Figure 1.1 DAS-II Clusters by Battery

THEORETICAL UNDERPINNINGS

The DAS-II was not developed solely to reflect a single model of cognitive abilities but was designed to address processes that often underlie children’s difficulties in learning and what scientists know about neurological structures underlying these abilities. The selection of the abilities to be measured by the DAS-II was influenced by a variety of theoretical points of view, but the end result is consistent with Gf-Gc theory (now commonly referred to as the Cattell-Horn-Carroll theory, or simply CHC). This is probably the best known and most widely accepted theory of intellectual factors among practitioners of individual psychological assessment and is derived from the Horn-Cattell Gf-Gc model [e.g., Cattell (1941, 1971, 1987), Cattell & Horn (1978), Horn (1988, 1991), Horn & Noll (1997)]. Gf and Gc refer, respectively, to “fluid” and “crystallized” intelligence, but current versions of the theory recognize as many as seven different broad cognitive factors or abilities. See Carroll (1993); Flanagan and McGrew (1997); Flanagan, McGrew, and Ortiz (2000); Flanagan and Ortiz (2001); Flanagan, Ortiz, and Alfonso (2007); Flanagan, Ortiz, Alfonso, and Mascolo (2002); Horn (1985, 1988, 1991); Horn and Cattell (1966); Horn and Noll (1997); McGrew (1997); McGrew and Flanagan (1998); Woodcock (1990); and Woodcock and Mather (1989) for discussions of Gf-Gc, now usually called the Cattell-Horn-Carroll (CHC) theory. Carroll’s monumental (1993) review and re-analysis of hundreds of factor analytic studies of many psychological tests provided a solid empirical foundation for CHC theory. The factor structure that Carroll devised on the basis of his research was remarkably congruent with the theoretical structure developed by Cattell and Horn (1978; Horn, 1988, 1991), which lent further credence to the amalgamated CHC theory as subsequently developed by Woodcock, McGrew, Flanagan, and others [e.g., Flanagan & McGrew (1997); Flanagan, McGrew, & Ortiz (2000); Flanagan & Ortiz (2001); Flanagan, Ortiz, & Alfonso (2007); Flanagan, Ortiz, Alfonso, & Mascolo (2002); Horn (1991); McGrew (1997); McGrew & Flanagan (1998); McGrew, Werder, & Woodcock (1991); Woodcock (1990, 1993); and Woodcock & Mather (1989)]. However, even with a growing consensus as to the nature and structure of human cognitive abilities, there remains substantive debate regarding the number of factors representing independent abilities in a cognitive model, the precise nature of each of those factors (Horn & Blankson, 2005; Carroll, 2005), and to what extent, if any, subtests from different test batteries that purport to measure a given factor actually do so (Alfonso, Flanagan, & Radwan, 2005).

Despite the fact that no single theory or model has universal acceptance, there is a common core of theory and research that supported the development of the DAS-II. Such research indicates that human abilities are complex and often are not best explained solely in terms of a single cognitive factor (g), or even in terms of several lower-order factors. These abilities are presented as multiple dimensions on which individuals show reliably observable differences, and are related to how children learn, achieve, and solve problems. Although these abilities are interrelated, they do not completely overlap, thus making many of them distinct (Carroll, 1993). The wide range of human abilities represents a number of interlinked subsystems of information processing that have structural correlates in the central nervous system, in which some functions are distinct and others are integrated. Some formulations of CHC theory (e.g., Carroll, 1993, 2005) include an overarching, single factor, g , at the top of the hierarchy. Others (e.g., Horn, 1991; Horn & Blankson, 2005) dispute the importance, or even the existence, of a single, overall level of cognitive ability and emphasize the importance of the separate abilities. Yet others (e.g., Flanagan & McGrew, 1997; Flanagan, McGrew, & Ortiz, 2000) do not take a rigid stand on the question of an overall g, but operationalize the theory on the basis of the separate factors. All of these versions of CHC theory maintain at least two strata of abilities: several broad abilities each including several narrow abilities. In the three-stratum model (e.g., Carroll, 2005), the narrow abilities are called Stratum I, the broad abilities Stratum II, and g, at the top of the hierarchy, Stratum III.

Flanagan and McGrew (1997); Flanagan, McGrew, and Ortiz (2000); Flanagan and Ortiz (2001); Flanagan, Ortiz, and Alfonso (2007); Flanagan, Ortiz, Alfonso, and Mascolo (2002); Horn (1991); McGrew (1997); McGrew and Flanagan (1998); McGrew, Werder, and Woodcock (1991); Woodcock (1990, 1993); and Woodcock and Mather (1989) have adopted a notation system, largely based on that of Carroll (1993). Symbols for broad (Stratum II) abilities are written with a capital G and italicized, lowercase letters (e.g., Ga is auditory processing, and Glr is long-term storage and retrieval). Symbols for narrow (Stratum I) abilities within the various broad abilities are usually written with one or two capital letters or a capital letter and a digit (e.g., SR is spatial relations within Gv, I is induction within Gf, and K1 is general science information within Gc). Other notations are used occasionally (e.g., PC:A and PC:S are, respectively, phonetic coding: analysis and phonetic coding: synthesis). Several similar, but not identical, verbal labels are given to the abilities (e.g., Gv has been called “visual processing,” “visual/spatial processing,” and “visual/spatial thinking”), so the more-or-less agreed-upon symbols function as a valuable common notation with less risk of misunderstanding.

The following section outlines some links between the DAS-II ability constructs and neuropsychological structures in the areas of verbal and spatial abilities, fluid reasoning abilities, several aspects of memory, and processing speed.

Broad Verbal and Spatial Abilities

The DAS-II Verbal and Spatial ability clusters reflect major systems through which individuals receive, perceive, remember, and process information. Both systems are linked to auditory and visual modalities and factorially represent verbal [crystallized intelligence (Gc)] and visual [visual-spatial (Gv)] thinking.

Neuropsychologically, there is strong evidence for the existence of these systems. They tend to be localized in the left and right cerebral hemispheres, respectively, although the localization is complicated (see, for example, Hale & Fiorello, 2004, pp. 67-78) and there are individual differences in areas of localization of function. Moreover, the systems are doubly dissociated—that is, they represent two distinct, independent systems of information processing (McCarthy & Warrington, 1990; Springer & Deutsch, 1989). The systems are independent insofar as each one may remain intact if the other is damaged. In the DAS-II, the two factors (verbal and spatial) are measured by the Verbal and Spatial clusters in both the Early Years and School-Age batteries.

Crystallized ability (Gc) refers to the application of acquired knowledge and learned skills to answering questions and solving problems presenting at least broadly familiar materials and processes. Virtually all tests of Gc are verbal, as that is the nature of many crystallized tasks: language is the primary means by which we express and use acquired knowledge. Most verbal subtests of intelligence scales primarily involve crystallized intelligence. Subtests of general knowledge and vocabulary are relatively pure measures of crystallized intelligence. The overlap between crystallized intelligence and verbal information processing is indeed so strong that we believe that the meaning of the factor and the test scores that measure it is best expressed as “Verbal,” as in the DAS-II cluster score.

We note here that within the area of auditory-verbal processing there are distinctions that have to be made between different types of cognitive processes. Most of the tasks that are included under the Gc factor are concerned with verbal knowledge (including vocabulary), comprehension of single or multiple sentences, and verbal reasoning. All these are relatively high-level cognitive tasks, requiring complex processing, analysis of meaning, and retrieval of information that has been stored in long-term verbal memory. In contrast, there are other verbal factors that require immediate, less complex verbal processing. Auditory short-term memory (Gsm) is measured by tasks that entail repeating words that have been heard, with little or no processing of the meaning of the words themselves. We can characterize this as relatively simple information processing. Similarly, auditory processing ability (Ga) is measured by tasks that require the individual to analyze the component sounds of words that are presented. Again, such tasks do not require the meaning of those words to be an important component of the task. Both Gsm and Ga will be discussed below.

Visual-spatial thinking (Gv) involves a range of visual processes, ranging from fairly simple visual perceptual tasks to higher level, visual, cognitive processes. Woodcock and Mather (1989) define Gv in part: “In Horn-Cattell theory, ‘broad visualization’ requires fluent thinking with stimuli that are visual in the mind’s eye. . . .” Although Gv tasks are often complex and mentally challenging, Gv primarily relies on visual processing that involves the perception of and ability to visualize mental rotations and reversals of visual figures. It is not dependent on the ability of the individual to use internal verbal language to help solve problems.

Again, we note at this point that not all “nonverbal” tasks measure Gv. Because we have stipulated the condition (which is borne out by factor-analytic research) that Gv tasks are not dependent upon the ability of the individual to use internal language in solving a problem, it follows that tasks that require this are measuring a different cognitive process. Gv tasks do not include the aspect of dealing with novel stimuli or applying novel mental processes, or using internal language to reason out the solution to a visually-presented problem, all of which characterize Gf tasks. This will be discussed below in the section on Integration of Complex Information Processing.

Auditory Processing Ability: Is it a Component of Verbal Ability?

It should be noted that Horn and Carroll both accepted that there is a separate factor of auditory processing (Ga) that is distinct from the verbal or Gc information processing system. Auditory processing is concerned with the analysis of sound patterns such as in speech sounds, rhythm, and sequences of sounds (Carroll, 2005; Horn & Blankson, 2005). Auditory processing ability is certainly related to the development of complex higher-order language skills. It is necessary but not sufficient for language development. It seems reasonable to suppose that auditory processing is mediated by a separate processing system that handles the analysis of auditory sensory input, and because of this, children with hearing impairment are likely to have difficulties with Ga tasks.

In the DAS-II, auditory processing (Ga) is measured by the Phonological Processing subtest, comprising four distinct components: Rhyming, Blending, Deletion, and Phoneme Identification and Segmentation.

Integration of Complex Information Processing

For normal cognitive functioning, the verbal and visual-spatial abilities operate as an integrated information processing system that is necessary for complex mental activity. Factorially, this integrative system is represented by the fluid reasoning (Gf ) ability. Fluid reasoning refers to inductive and deductive reasoning, presenting problems that are new to the person doing the reasoning. The vast majority of fluid reasoning tests use nonverbal (that is, visual) stimuli using pictures or figures. These require an integration of verbal and nonverbal thinking. Indeed, it seems likely that the best measures of Gf always require integrated analysis of both verbal and visual information. This is achieved through the presentation of visual problems that, for most efficient solution, require the individual (1) to encode the components of the visual stimulus, (2) to use internal language to generate hypotheses, (3) to test the hypotheses, and (4) to identify the correct solution.

Neuropsychologically, it seems that the integrative function of frontal lobe systems is central to executive function, which is involved in planning and other complex mental processes (Hale & Fiorello, 2004, pp. 64-67; Luria, 1973; discussed by McCarthy & Warrington, 1990, pp. 343-364), and it is therefore reasonable to hypothesize that it may provide a structural correlate for Gf. Similarly, it is clear that the corpus callosum has a major role in connecting the right and left cerebral hemispheres, and that limitations in callosal transmission may be implicated in cases of poor visual-verbal integration. Whatever the localization of specific mechanisms may be, the fact that our brains have an integrative function seems incontrovertible. The best tests of Gf require that integrative process.

In the DAS-II, the Gf factor is measured in the Upper Early Years and School-Age batteries by the Nonverbal Reasoning cluster.¹ The subtests measuring this ability require integrated analysis and complex transformation of both visual and verbal information, and verbal mediation is critical for the solution of these visually presented problems for most individuals.

Short-Term Memory (Verbal and Visual) Systems

Short-term memory (Gsm) refers to one’s ability to apprehend and maintain awareness of elements of information for events that occurred in the last minute or so. Gsm refers to aspects of memory that have limited capacity and that lose information quickly unless an individual activates other cognitive resources to maintain the information in immediate awareness. CHC theory does not distinguish, at the second-order, group factor level, between separate, modality-related visual and verbal memory systems. At the broad factor level there is only a single short-term memory factor (Gsm) that should really be called auditory short-term memory.

Because of evidence from both cognitive psychology and neuropsychology that shows clearly that verbal and visual short-term memory systems are distinct and independent (Hitch, Halliday, Schaafstal, & Schraagen, 1988; McCarthy & Warrington, 1990, pp. 275-295), the DAS-II does not treat short-term memory as unitary but keeps auditory and visual short-term memory tasks as distinct measures. Additionally, several subtests combine to create a working memory (Gsm MW) factor that is separate from auditory short-term memory (Gsm MS), as measured by the Recall of Digits Forward subtest, and the visual short-term memory (Gv MV) abilities measured by the Recall of Designs and Recognition of Pictures subtests.

Integration of Verbal and Visual Memory Systems

The long-term storage and retrieval (Glr) factor in the CHC model is typically measured by tests that have both visual and verbal components. Long-term storage and retrieval ability involves memory storage and retrieval over longer periods of time than Gsm. How much longer varies from task to task, but it is typically of the order of 1 to 30 minutes.

McCarthy and Warrington (1990, p. 283) call this “visual-verbal” short-term memory and conclude that it is underpinned by another distinct and independent, dissociable information-processing system. While its relationship with other processes is relatively small, it may be an important type of “gateway” process underlying some types of working memory. Holding information in visual-verbal short-term memory may be necessary in order to solve problems that require the manipulation and transformation of visual information that can be labeled verbally.

In the DAS-II, the visual-verbal memory factor (Glr) is measured by the Recall of Objects subtest. In this task, an array of pictures is presented, but they have to be recalled verbally. Sequential order is not important, and the child is able to organize and associate pictures in any way that helps in remembering them.

Processing Speed

The DAS-II Processing Speed cluster measures the CHC processing speed factor (Gs). This factor refers to the ability to automatically and fluently perform relatively easy or over-learned cognitive tasks, especially when high mental efficiency (i.e., attention and focused concentration) is required. It is typically measured by tests that require relatively simple operations that must be performed quickly—speed of decision, speed of naming, clerical speed, and so on. These types of timed activities are more complex than those involved in simple reaction-time paradigms, which seem to form their own factor (Gt), a factor not assessed by the DAS-II, nor by most cognitive ability tests.

While individual differences in neural speed may be one of the determinants of performance on processing speed tasks, it is clear that other determinants are involved. Speed of response may reflect not only neural speed but also perhaps efficiency in accessing information, efficiency in holding information in short-term memory, efficiency in visual-verbal integration, and willingness to commit to a decision and threshold for doing so. Performance on Gs tasks is not easily improved with practice. Prior experience on similar tasks is unlikely to be helpful. Therefore, measures on such tasks do reflect some function of the underlying speed and efficiency of processing systems.

DESCRIPTION OF DAS-II

The Differential Ability Scales—Second Edition (DAS-II; Elliott, 2007a) is an individually administered battery of cognitive tests for children and adolescents aged 2 years, 6 months (2:6) through 17 years, 11 months (17:11). Because the DAS-II covers such a wide age range, it is divided into three levels: Lower Early Years (ages 2:6 through 3:5); Upper Early Years (normally covering ages 3:6 through 6:11, but normed through 8:11); and School-Age (normally covering ages 7:0 through 17:11, but also normed for ages 5:0 through 6:11). The three levels allow both items and clusters that are appropriate to the several age ranges. It was designed to measure specific, definable abilities and to provide reliable, interpretable profiles of strengths and weaknesses. These profiles may lead to individualized interventions or treatments for students with learning concerns or issues. The DAS-II is considered suitable for use in any setting in which the cognitive abilities of children and adolescents are to be evaluated, although several of the DAS-II subtests may not be appropriate for students with severe sensory or motor disabilities. The DAS-II cognitive battery yields a composite score labeled General Conceptual Ability (GCA) that is a measure of psychometric g, defined as “the general ability of an individual to perform complex mental processing that involves conceptualization and transformation of information” (Elliott, 2007b, p. 17).

CAUTION

Several of the DAS-II subtests may not be appropriate for students with severe sensory or motor disabilities.

Organization of the DAS-II

The DAS-II contains a total of 20 subtests grouped into Core or Diagnostic subtests. The Core subtests are those used to compute the GCA and three cluster scores: Verbal Ability, Nonverbal Reasoning Ability, and Spatial Ability. The Diagnostic subtests measure aspects of memory, speed of processing and early concepts taught in schools. They yield three cluster scores: Processing Speed, Working Memory, and School Readiness. These diagnostic subtests are considered important and useful in the interpretation of an individual’s strengths and weaknesses in information processing, but they do not contaminate the GCA with subtests that have low g loadings.

This separation of Core and Diagnostic subtests is one of the strengths of the DAS-II. For a point of comparison, the Wechsler Intelligence Scale for Children, 4th ed. (WISC-IV; Wechsler, 2003) excludes the Information, Word Reasoning, Arithmetic, Picture Completion, and Cancellation subtests from the FSIQ and Indices, but does include in the IQs subtests such as Coding and Symbol Search, which are not good measures of complex mental processing or intellectual ability ( g ). The Stanford-Binet Intelligence Scale, 5th ed. (SB5; Roid, 2003) includes all subtests in the total score. The Woodcock-Johnson III Cognitive battery (WJ III; Woodcock, McGrew, & Mather, 2001) includes low-g-loading tests, but only in proportion to their g loading

DON’T FORGET

The separation of the DAS-II into Core and Diagnostic subtests can be helpful in reducing the overall administration time and a student’s fatigue since examiners can tailor their assessments, administering only those subtests that are relevant based on the specific and different referral questions.

The Lower Early Years battery of the DAS-II consists of four core subtests that combine to yield the GCA and three diagnostic subtests that may be administered. The Upper Early Years battery includes six core subtests and an additional 11 optional diagnostic subtests. The School-Age battery includes six core subtests and nine additional diagnostic subtests. Some of the Early Years subtests can also be used at the school-age level, especially at younger ages, for diagnostic purposes. For the Upper Early Years and the School-Age batteries, the subtests not only combine to produce the GCA but also yield five or six cluster scores. For Upper Early Years children, these cluster scores represent Verbal (Gc), Nonverbal Reasoning (Gf ), and Spatial (Gv) abilities along with School Readiness, Working Memory (Gsm), and Processing Speed (Gs). For School-Age children, the cluster scores represent Verbal (Gc), Nonverbal Reasoning [(Gf ) fluid reasoning (Keith, 1990)], and Spatial (Gv) abilities along with Working Memory (Gsm) and Processing Speed (Gs) (see Rapid Reference 1.2 and Figure 1.1). Although the “typical” Upper Early Years battery is given to children aged 3 years, 6 months through 6 years, 11 months and the “typical” School-Age battery to children 7 years, 0 months through 17 years, 11 months, the Upper Early Years and School-Age batteries were also normed for an overlapping age range (5 years, 0 months through 8 years, 11 months).

Normative Overlaps

Depending on the examinee’s age, if an examinee of low ability has little success at the ages covered by the battery you initially selected, you may be able to administer subtests from a lower level of the test. Conversely, if an examinee has high ability and has few failures at the ages covered by the battery you initially selected, you can administer subtests from a higher level of the test. All subtests at the Upper Early Years and School-Age Level have overlapping normative data for children ages 5: 0 to 8:11. This overlap provides the examiner flexibility when testing bright younger children or less able older children. In these cases, subtests appropriate for the individual’s abilities are available. For example, the Upper Early Years subtests can be administered to children ages 6:0 to 8:11 for whom the School-Age Level is too difficult. Similarly, the School-Age subtests can be administered to children ages 5: 0 to 6:11 for whom the Upper Early Years is insufficiently challenging. In such cases, the examinee’s raw scores can be converted to ability scores and then to T scores in the normal way. For children in the overlapping age range, examiners may choose to give either battery or choose one battery and administer additional subtests from the other battery.

DON’T FORGET

If a student has little success at the ages covered by the battery you initially selected, you may be able to administer subtests from a lower level of the test.

Changes from DAS to DAS-II

Several goals were accomplished with the revision of the DAS to the DAS-II. Rapid Reference 1.3 lists the key features that were accomplished and changes made for this second edition.

In the DAS-II, many of the core subtests will be recognizable to DAS examiners, but there have been significant changes and modifications to some. For example, Block Building and Pattern Construction have been combined into one subtest; Recall of Digits has been expanded to two subtests: Recall of Digits- Forward and Recall of Digits-Backward; and Early Number Concepts has been removed from the GCA and is now included in the School Readiness cluster. There are four new diagnostic subtests (Phonological Processing, Recall of Digits-Backward, Recall of Sequential Order, Rapid Naming). The major structural changes in the DAS-II are the inclusion of separate Nonverbal Reasoning and Spatial Ability clusters at the Upper Early Years and the creation of three new clusters (Working Memory, Processing Speed, School Readiness), developed to help examiners assess the skills of the child.

Rapid Reference 1.3

DAS-II Key Revisions

• Updating of norms

• CHC interpretative basis now noted explicitly in manual and record form

• Development of three new Diagnostic Clusters (Working Memory, Processing Speed, School Readiness)

• Addition of four new subtests (Phonological Processing, Recall of Digits Backward, Recall of Sequential Order, Rapid Naming)

• Downward extension of Matrices subtest to age 3 years, 6 months, enabling the Nonverbal Reasoning cluster to be measured at the Early Years level.

• Core cluster scores (Verbal, Nonverbal Reasoning, Spatial) are now the same throughout the age range from 3:6 through 17:11

• Block Building and Pattern Construction combined into one subtest

• Revising content of 13 subtests

• Updating artwork

• Eliminating three achievement tests

• Linking DAS-II to the WIAT-II and providing correlational data also for the K-TEA-II and the WJ-III Achievement batteries

• Providing Spanish translation for nonverbal subtests

• Providing American Sign Language translation for nonverbal subtests in every kit for use by, and the training of, interpreters

• Publishing with Scoring Assistant computer software

Rapid Reference 1.4 compares the number of items on the DAS and DAS-II and the number of items retained and added. The DAS-II has increased the number of items on five of the core tests, and two of the diagnostic tests and decreased the number on two subtests. The greatest increase in items came on the Matrices subtest (35 items to 56 items, a 60 percent increase) while the largest decrease came on Word Definitions (42 items decreased to 35, a 17 percent decrease). The regionally problematic word “wicked” was removed from Word Definitions. Four subtests (Recognition of Pictures, Recall of Objects, Matching Letter-Like Forms, and Speed of Information Processing) remain exactly the same on the DAS-II.

Rapid Reference 1.4

All DAS-II subtests have also been aligned with Cattell-Horn-Carroll (CHC) abilities (see Rapid Reference 1.5). This allows the examiner to use commonly understood and agreed-upon terminology when interpreting what the DAS-II is measuring. CHC theory provides for the interpretation of both Broad and Narrow abilities. The DAS-II provides measures of each of the seven most robust and replicable factors derived from research.

Rapid Reference 1.5

Wider Score Ranges

The DAS-II has a wider range of possible T scores for the subtests and Standard Scores for the clusters in comparison with the DAS first edition. In the DAS, T scores ranged from 30 to 70 (that is, two standard deviations (SDs) on either side of the mean of 50), whereas in the DAS-II the range is 20 to 80 (three SDs on either side of the mean). Similarly, for the GCA, SNC and the cluster scores, the maximum DAS range was 45 to 165, whereas in the DAS-II the maximum range is 30 to 170 (that is, 4.67 SDs on either side of the mean).

RELATIONSHIPS BETWEEN THE DAS AND THE DAS-II

Rapid Reference 1.6 provides the results of comparisons of scores obtained on the first-edition DAS and the DAS-II. The major study presented in the DAS-II Introductory and Technical Handbook gave children the two batteries with a short interval between tests. We also present a clinical study carried out on children identified as ADHD in which the assessments were carried out over a period of years.

Over Short Periods of Time

The relationship between the DAS and the DAS-II was examined in a sample of 313 children aged 2:6 to 17:11 (Elliott, 2007b). Each test was administered in counterbalanced order with 6 to 68 days between testing. The overall correlation coefficients show that the Verbal Ability scores for the DAS and the DAS-II were the most highly related (r = .84) followed by the GCA (r = .81) and the Special Nonverbal composite (r = .78). As shown in Rapid Reference 1.6, the average DAS-II GCA is 2.7 points lower than the GCA of the DAS. The difference between the two tests is small for the Verbal Ability (0.1 points), while the Nonverbal Reasoning and Spatial abilities differ by 4 to 5 points. These differences, both in size and direction, are generally somewhat lower than expected according to the Flynn Effect (Flynn, 1984, 1987, 1998). The results indicate that if examinees continue to be assessed using the first edition of the DAS, their scores may be inflated by up to 4 or 5 standard score points in comparison with the DAS-II.

Over Long Periods of Time

In a small sample (N = 26) of children with ADHD who were administered the DAS first and then, after 3 to 6 years, were given the DAS-II, small changes in test scores were observed (Schlachter, Dumont, & Willis, unpublished manuscript). In almost all cases, the test scores on the DAS-II were lower than their earlier scores on the original DAS. Only Matrices and Recall of Objects-Immediate were higher on the DAS-II, and in each case by less than 1 point. The smallest mean difference in composite scores was shown by the Nonverbal Reasoning cluster, with a mean score on the DAS-II 1.8 points lower than that on the DAS. The greatest difference in composite scores was shown by the GCA, with the mean score on the DAS-II being 3.9 points lower than that on the DAS. For individual subtests, the largest change between the DAS and the DAS-II was on Verbal Similarities and Recall of Designs (-4.8 and -3.9 points, respectively).

Rapid Reference 1.6

STANDARDIZATION AND PSYCHOMETRIC PROPERTIES

The DAS-II was standardized and normed on 3,480 children selected to be representative of non-institutionalized, English-proficient children aged 2 years 6 months through 17 years 11 months living in the United States during the period of data collection (2005). Although the DAS-II standardization excluded those children with severe disabilities (since for these children the DAS-II would be inappropriate), it did include children with mild perceptual, speech, and motor impairments, if the examiner judged that the impairments did not prevent the valid administration of the test. The demographic characteristics used to obtain a stratified sample were age, sex, race/ethnicity, parental educational level, and geographic region.

Additional samples of children, ranging in size from 54 to 313, were tested during standardization with three additional cognitive measures, three achievement measures, and two measures of school readiness, to provide evidence of validity. These additional children were not included in the norms calculation.

For the category of race/ethnicity, individuals were classified as White (N = 2,176), African American (N = 538), Hispanic American (N = 595), Asian (N = 137) and Other (N = 34). The five parental education categories ranged from one to eight years of education to 16 or more years of education. The four geographic regions sampled were Northeast, Midwest, South, and West. Demographic characteristics were compared to the October 2005 U.S. Census populations and were matched in three-way tables across categories and not just within single categories (i.e., age × race × parent education; age × sex × parent education; age × sex × race; and age × race × region). Total sample percentages of these categories and subcategories were very close to the Bureau of the Census data and seldom different by more than 1 percentage point.

In the standardization sample, there were 18 age groups: 2:6-2:11, 3:0-3:5, 3:6-3:11, 4:0-4:5, 4:6-4:11, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, and 17 years. In each six-month age group between 2 years 6 months and 4 years 11 months, there was a total of 176 children, while from ages 5 through 17 there were 200 children in each one-year age group. In each six-month age group between 2 years 6 months and 4 years 11 months, there were approximately equal numbers of males and females, while for all remaining age groups there were 100 males and 100 females per group. In our opinion, this sampling methodology was excellent.

RELIABILITY OF THE DAS-II

The DAS-II has excellent reliability (see Rapid Reference 1.7 for the average internal consistency reliability and standard error of measurement (SEm) for each Composite and Cluster). Average internal consistency reliability coefficients for the GCA and the Special Nonverbal Composites are above .90 for the Lower Early Years, Upper Early Years, and School-Age level. For the clusters, average internal consistency reliability coefficients for the Lower Early Years, Upper Early Years, and School-Age level are (a) .93, .89, and .89 for Verbal Ability (b) .87, .89,

Rapid Reference 1.7

and .92 for Nonverbal Reasoning Ability, (c) .95 for Spatial Ability, (d) .95 for Working Memory, (e) .90 for Processing Speed, and (f ) .90 for School Readiness. These numbers indicate that all of these overall cluster scores are “reliable,” the term “reliable” being defined by Sattler (2008) as a reliability coefficient with a value between .80 and .99.