Test and Measurement Chapter 10: The Wechsler Intelligence Scales ...

Report 6 Downloads 30 Views
Test and Measurement Chapter 10: The Wechsler Intelligence Scales: WAIS-IV, WISC-IV and WPPSI-III 

Throughout his career, Wechsler emphasized that factors other than intellectual ability are involved in intelligent behavior. Today, there are three Wechsler intelligence tests, the Wechsler Adult Intelligence Scale, Fourth Edition >n (WAIS-IV), the Wechsler Intelligence Scale for Children, Fourth Edition (WISC-IV), and the Wechsler Preschool and Primary Scale of Intelligence, Third Edition (WPPSI-III).

The Wechsler Intelligence Scales    

The role of nonintellective factors is apparent in the Wechsler intelligence scales. Wechsler’s test did not directly measure nonintellective factors; it took these factors into careful account in its underlying theory. Wechsler (1939) also correctly noted that the Binet scale emphasis on speed, with timed tasks scattered throughout the scale, tended to unduly handicap older adults. Wechsler criticized the then existing Binet

Point and performance scale concepts  



Differences between the Wechsler and the original Binet scales were (1) Wechsler’s use of the point scale concept rather than an age scale and (2) Wechsler’s inclusion of a nonverbal performance scale. The point scale concept o In a point scale, credits or points are assigned to each item. o . This scale makes it easy to group items of a particular content together, which is exactly what Wechsler did. o Similar concept was used in the 1986 Binet scale. o Yielded not only a total overall all score but also scores for each content area. The performance scale concept o Wechsler included an entire scale that provided a measure of nonverbal intelligence: a performance scale. o The performance scale consisted of tasks that require a subject to do something rather than merely answer questions o Original Wechsler scale, however, included two separate scales. o The verbal scale provided a measure of verbal intelligence, and the performance scale a measure of nonverbal intelligence. o Wechsler’s new scale was the first to offer the possibility of directly comparing an individual’s verbal and nonverbal intelligence that is, both the verbal and performance scales were standardized on the

same sample, and the results of both scales were expressed in comparable units. o They not only measure intelligence but also provide the clinician with a rich opportunity to observe behavior
 in a standard setting. From the Wechsler-Bellevue Intelligence Scale to the WAIS-IV  

First effort to measure adult intelligence was poorly standardized. Its normative sample consisted of a non-representative sample of 1081 whites from the eastern United States

Scales, Subtests and Indexes       

Wechsler defined intelligence as the capacity to act purposefully and to adapt to the environment. Wechsler’s definition implies that intelligence comprises several specific interrelated functions or elements and that general intelligence results from the interplay of these elements. In the WAIS-IV, Wechsler’s basic approach is maintained. First, there are individual subtests, each of which is related to a basic underlying skill or ability Each of the various subtests is also part of a broader “index.” WAIS-IV, the subtests are sorted into four indexes: verbal comprehension, perceptual reasoning, working memory, and processing speed. An index is created where two or more subtests are related to a basic underlying skill

A Closer Look as Subtests 



The vocabulary subtest o Ability to define words o If an individual has shown deterioration i.e., lowered performance compared with a previously higher level) because of emotional factors or brain damage, for example, vocabulary is one of the last functions to be affected. o Mild concentration difficulties lower optimal performance on arithmetic and digit span tasks, such difficulties generally do not affect vocabulary until they become quite severe o Can use it to evaluate baseline or premorbid intelligence The similarities Subtest o Consists of paired items of increasing difficulty. o The subject must identify the similarity between the items in each pair. o This subtest measures the subject’s ability to see the similarity between apparently dissimilar objects or things.

o Individuals with schizophrenia tend to give idiosyncratic concepts, or concepts that have meaning only to them. 











The arithmetic Subtest o Contains approximately 15 relatively simple problems in increasing order of difficulty. The ninth most difficult item is as easy as this:  “A person with $28.00 spends $.50. How much does he have left? The digit span subtest o Requires the subject to repeat digits, given at the rate of one per second, forward and backward o Measures short-term auditory memory and is one of the core subtests in the working memory index o Non-intellective factors often influence the results The Information Subtest o College students typically find the information subtest relatively easy and fun o Items appear in order of increasing difficulty. o The information subtest involves both intellective and nonintellective components, o Factors such as curiosity and interest in the acquisition of knowledge tend influence test scores. The comprehension Subtest o Has three types of questions. o The first asks the subject what should be done in a given situation, as in, “What should you do if you find an injured person lying in the street?’ o The second type of question asks the subject to provide a logical explanation for some rule or phenomenon, as in, why do we bury the dead? o He third type asks the subject to define proverbs such as, “A journey of 1000 miles begins with the first step.’ o Measures common sense. o The person’s emotional disturbance interferes with his or her judgment and results in an inappropriate response.  With the example what do you do if you find someone hurt on the side of the road The Letter Number Sequencing Subtest o Not required to obtain an index score o It is made up of items in which the individual is asked to reorder lists of numbers and letters. For example, Z, 3, B, 1, 2, A, would be reordered as 1, 2, 3, A, B, Z o Elated to working memory and attention The digit Symbol-Coding subtest o Quires the subject to copy symbols







o After completing a short practice sample, the subject has 120 seconds to copy as many symbols as possibly o The subtest measures the ability to learn unfamiliar tasks, visual motor dexterity, degree of persistence and speed of performance The block design subtest o Included in nonverbal measures of intelligence o Materials for the block design subtest include nine variously colored blocks. o The materials also include a booklet with pictures of the blocks arranged according to a specific geometric design or configuration o Must arrange the blocks to reproduce increasingly difficult designs. o The subtest provides an excellent measure of nonverbal concept formation, abstract thinking, and neurocognitive re impairment o It is one of the core measures of the perceptual reasoning index scale in the WAIS-IW. The Matrix Reasoning Subtest o Was included in the WAIS-IV as a core subtest in the perceptual reasoning index scale in in an e: effort to enhance the assessment of fluid intelligence, which involves our ability to reason. o Subject is presented with non-verbal, figural stimuli o The task is to identify a pattern or relationship between the stimuli. The symbol Search Subtest o I snow a core measure in the processing speed index scale o Shown two target geometric figures. o He task is then to search from a long a set of five additional search figures and determine whether the target appears in the search group.

From Raw Scores to Scaled and index scaled scores     



Together each subtest produces a raw score—that is, a total number of points—and has a different maximum total. Standard deviation converted to standard or scaled scores with a mean of 10 and a standard deviation of 3. The test developers used a statistical method called inferential norming These create subject group norms that can be used to compare A variety of statistical indexes or “moments,” such as means and standard deviations, were calculated for each of the 13 age groups of the stratified normative sample (These were plotted across age groups to derive estimates of age group midpoint population moments. In the end, the test developers were able to derive reference-group norms, which allow the test user to compare subjects at the subtest level. As will be seen later, such reference group norms can be extremely valuable in interpreting scores. The four composite index scales are then derived by summing the core subtest scores

Index Scores     

Verbal comprehension index might best be thought of as a measure of crystallized intelligence. This index is as a “more refined,” “purer” measure of verbal comprehension than is the VIQ because it excludes the arithmetic and digit span subtests, which have 2 attentional or working memory components The perceptual reasoning index, consisting o of visual puzzles, block design believed to be a measure of fluid intelligence. Working memory refers to the information that we actively hold in our minds, in contrast to our stored knowledge, or long-term memory The processing speed index attempts to measure how quickly your mind works.

FSIQ’s  

It is obtained by summing the age-corrected scaled scores all four-index composites. Again, a deviation IQ_ with a mean of 100 and a standard deviation of 15 is obtained. The FSIQ represents a measure of general intelligence.

Interpretive Features of the Wechsler Tests 

Index Comparisons o You are comparing the PIQ with the VIQ- if they match each other say they are both 60 then you can confirm that a person is mentally retarded, but what if one exceeds the other o This may mean that there are other factors like culture that could be playing a role o There are limits to index comparisons o Experiments showed that all different types of people got the same IQ but all differed in VIQ’s and PIQ’s

Pattern Analysis      

Evaluates relatively large differences between subtest scaled scores Peoples emotions will effect there performance on different tasks Provided a host of patterns tentatively proposed as diagnostically significant Analysis of patterns must be done consciously Should be used to generate a hypothesis ^Then must be proven by other pieces of data

HYPOTHETICAL CASE STUDIES ON PAGE 216

Psychometric properties of the Wechsler Adult Scale  



Standardization o 2200 adults, divided into 13 age groups from 16-90 as well as thirteen specialty groups Reliability o Split-half method: .98 o For the FSIQ: .96 o For the VIQ .95 o For the PIQ .94 o We have more confidence in the verbal comprehension -95% confidence interval Validity o There are considered valid

Evolution of the Wechsler Adult Scales  Is extensively used, is very reliable Downward extensions of the WAIS-III and the WISC-IV and the WPPSI-II   

WISC-IV was first published in 1949 and revised in 1974 and 1991 and then again in 2003 It measures intelligence from ages 6 to 16 years 11 months The WPPSI-II measures from2.5-7 years three months

The WISC-IV        

Provides a composite index The original form of this scale was based off form 11 of the WechslerBellevue scale Contains 15 subtests Three subtests used in earlier versions—picture arrangement, object assembly, and mazes—were entirely deleted. The WISC-IV al abandoned the VIQ-PIQ dichotomy. There is an emphasis on the modern a cognitive psychology concept of working memory and processing speed. GO OVER THE SPECIFIC SUBTESTS Item bias o Important innovations in the W1SC-IV is the use of empirical data to identify item biases. o Use trained judges to examine item content. The use of item analysis and other statistical procedures to determine which items may be biased adds an important and necessary dimension in the elimination of bias













o That major reason that empirical approaches are needed is that expert judgment judgments simply do not eliminate differential performance as a function of gender, race, ethnicity, and socioeconomic status. o Has not eliminated item bias, Standardization o Sample consisted of 2200 children selected to represent the March 2000 U.S. census. o The sample contained 100 boys and 100 girls at each age from 6 through 16 years Raw scores and scaled scores o Scaled scores are calculated from raw scores on the basis of norms at each age level, just as in the WAIS-IV.’ o The mean scaled score is also set at 10 and the standard deviation at 3. o Scaled scores are summed for the four indexes and IQs. These totals are then compared against a single table of standard scores with a mean of 100 and a standard deviation of 15 for each of the indexes and FSIQ Interpretation o Evaluating each of the four major indexes to examine for deficits in any given area and evaluate the validity of the FSIQ o Validity falls into question when there are large discrepancies Reliability o Split-half reliabilities for the WISCTVs composites range from .88 for processing speed to .97 for the FSIQ o Naturally, reliabilities for individual subtests run lower, o Test-retest reliability coefficients run a bit below those obtained id using the split-half method. WISC-IV Validity o Examines all of the relevant evidence that indicates whether a test score measures what it purports to o Bb
 problems have been confirmed by more recent analyses o No significant mean score differences across the cultures on the WISCIII subtests, FSIQ, VIQ, PIQ, or index scores, although mean scores; were effected by affluence and education o Results supported measurement invariance across normative and clinical samples o There is much evidence of the validity of the WISC-IV. The WPPSI-III o Scale for children 4 to 6 years of age, the WPPSI. o . Only two unique subtests are included: (1) animal pegs, an optional test that is timed and requires the child to place a colored cylinder into an appropriate hole in front of the picture of an animal; and (2)

o o o o o o o

sentences, an optional test of immediate recall in which the child is asked to repeat sentences presented orally by the examiner. The WPPSI-III was substantially revised It contains most of the new features of the WISC-IV, including five composites, but did not delete the VIQ_0r PIQ. Can now be used for children as young as 2 years, 6 months, and d has a s a special set of subtests specifically for the age range of 2 to 3 years, 11 months Made compatible with the Adaptive Behavior Assessment System, the WIAT-II, and the Differential Abilities S Scale. New subtests were added to enhance the measurement of fluid reasoning, p processing speed, and receptive, expressive vocabulary. Includes updated norms stratified as in WISC-IV and based on the October 2000 census. . It was first published in 1967 and revised in 1989 and again in 2002. Its changes and innovations parallel those found in the WISC-IV.