Hayward, Stewart, Phillips, Norris, & Lovell 4

At-a-Glance Test Review: Woodcock-Johnson Tests of Achievement III (WJ-III)

Name of Test: Woodcock-Johnson Tests of Achievement III (WJ-III)
Author(s): Woodcock, McGrew, and Mather
Publisher/Year: Riverside Publishing, 1989, 2001
Forms: parallel forms A and B
Age Range: 2 years to 90 years
Norming Sample: The norming study was conducted over the period from 1996 to 1999. The authors co-normed WJIII COG and WJIII ACH. Total Number: 8, 818, Number and Age: The sample was comprised of preschool (n=1 143), school age kindergarten though Grade 12, and adults (n=1 843). Location: Four major geographic U.S. regions, Demographics: Geographic region, gender, race, ethnicity, community size, and education or employment status were reported. Rural/Urban: Communities were reported by community size. SES: not provided, Other (Please Specify): Adult demographics were reported by employment, education, and occupational levels.
Summary Prepared By (Name and Date): Eleanor Stewart October and December 2007
Test Description/Overview:
The Woodcock-Johnson III Test of Achievement consists of a comprehensive set of subtests that aim to assess the student’s academic achievement and cognitive abilities. To do so, the authors have assembled an array of tasks that tap a range of skills that contribute to academic achievement. The WJIII is comprised of two separate batteries, the WJIII Test of Cognitive Abilities WJIII COG) and the WJIII Tests of Achievement (WJIII ACH). Each is further divided into Standard and Extended subtests. The WJIII ACH was available for review. It consists of 12 subtests and 10 extended subtests (see list below under Areas Tested).
Theory: The WJIII is based on the model proposed by Cattell, Horn, and Carroll. The CHC theory of cognitive abilities that states a single factor, General Intelligence, overarches eight broad ability domains. The Buros reviewer notes that while an abundance of empirical support exists for the CHC model, it is unclear how the model guided item development, particularly for the WJ-III ACH.
Purpose of Test: The purpose of the test is to assess academic strengths and weaknesses and to assess cognitive abilities.
Areas Tested: WJIII ACH consists of 12 subtests and 10 extended battery tests. The Standard Test consists of: letter-word identification, reading fluency, story recall, understanding directions, calculation, math fluency, spelling, writing fluency, passage comprehension, applied problems, writing samples, and story recall-delayed. Extended tests are: word attack, picture vocabulary, oral comprehension, editing, reading vocabulary A - synonyms, B - antonyms, C - analogies, quantitative concepts - A concepts, - B number series, academic knowledge - science, social studies, humanities, spelling of sounds, sound awareness - rhyming, deletion, substitution, reversals, punctuation and capitalization.
Areas Tested: Oral Language Vocabulary Narratives, Print Knowledge Alphabet Other (Please Specify) rebus, Phonological Awareness Segmenting Elision Rhyming Other (Please Specify) substitution, deletion, reversal, Reading Single Word Reading/Decoding Comprehension, Spelling, Writing Letter Formation Capitalization Punctuation Conventional Structures Word Choice Details, Listening Lexical
Who can Administer: The authors state that the examiner should have graduate level course and practicum experience.
Administration Time: Administration time varies according to which tests and clusters are administered.
Test Administration (General and Subtests): General administration is described in Chapter 3 of the examiner’s manual and in the test books at the beginning of each test. Icons on the tabs in the test books allow the examiner to quickly identify the presentation mode and materials needed. The authors list the tests which are appropriate for preschool children. These tests are: Letter-Word Identification, Story Recall, Understanding Directions, Spelling, Passage Comprehension, Applied Problems, Story Recall-Delayed, Word Attack, Picture Vocabulary, Oral Comprehension, Academic Knowledge, and Sound Awareness (p.16). They state, “ subjects of age two can generally perform the beginning tasks presented in Picture Vocabulary, Understanding Directions, or Academic Knowledge.” Tables at the beginning of each of these tests indicate the first item to administer for the preschool, kindergarten, and grade levels.
Test Interpretation: Raw scores are tabulated for each test and converted to standardized scores. The tests can be hand scored. The CD provides an interpretation of a range of scores available. Reports can be generated from the formats provided on the CD (English or Spanish). Clusters of tests include: Reading Clusters comprised of Broad Reading, Basic Reading, and Reading Comprehension; and Oral Language Clusters - Oral Language Standard.
Standardization: Age equivalent scores Grade equivalent scores Percentiles Standard scores Stanines
Other (Please Specify) The WJIII scoring system is said to provide any standardized score needed from the database provided.
N.B. The psychometric evidence for the WJIII is extensive. In order to present information that facilitates this review, I chose to use the summaries provided by the Buros reviewers as I felt that their comments were informative and comprehensive.
Reliability:
Internal consistency of items: The correlations across tests were high, .80 and above, clusters were in the .90 range and above. WJ-III ACH had no reliability coefficients below .70.
Test-retest: Selected subtests were reported. The first of two studies reported median reliabilities from the .70s to .90s. The second study used a one-year interval and found reliabilities in the .80s and .90s. Very high reliabilities are reported for Writing Sample, Writing Fluency, and Handwriting, which involve subjective ratings.
Inter-rater: Reviewer Sandoval writes: “Interrater reliabilities for the writing test evaluations based on correlations of scores are reported to be in the high .90s” (Cizek & Sandoval, 2003, p. 1026).
Other (Please Specify): Equivalence of forms: Data provided confirms that the two forms may be used interchangeably (McGrew & Woodcock, 2001, pp. 45-47).
Validity:
Content: The content of the WJ III is consistent with other achievement tests in use. The evidence from correlational analysis and confirmatory factor analyses presented demonstrates expected relationships among tests.
Criterion Prediction Validity: The evident for criterion prediction validity is laid out on pages 80 through 96 in the technical manual (McGrew & Woodcock, 2001), and compares selected tests and clusters against a range of measures including the OWLS, WRAT-III, KAIT, DRT, WAIS-III, WPPSI-R, WIAT, Leiter-R, KTEA, and CAS Scales. These tests represent a range of well-known cognitive and achievement tests. The section is extensive and the results are specific to each construct. “The cognitive clusters intended to predict achievement do correlate with the achievement clusters yielding correlations in the .70 range. These correlations are higher than those found between ability scores and achievement scores on many other cognitive measures suggesting excellent predictive validity” (Cizek & Sandoval, 2003, p. 1026).
Construct Identification Validity: Age and group (children with disabilities) validity is evidenced.
Differential Item Functioning: Data presented support CIBS-R in terms of the absence of bias across groups of interest. Details are somewhat lacking. Factor analysis provides evidence that the same factors are being measured across groups.
Other (Please Specify): Extensive information is provided in the technical manual.
Summary/Conclusions/Observations: This is a comprehensive test. I found the manuals difficult to wade through and I found myself having to return to some sections several times in order to complete my work. It became very difficult to summarize the information and for that reason I chose to refer to the Buros reviewers’ comments. Both of the Buros reviewers were impressed with the WJ-III.
Clinical/Diagnostic Usefulness: I think that the test items reviewed were probably too difficult for preschool and kindergarten children. With basals of 6 items on most tests, I doubt that young children would achieve success on most WJIII tests.
I think that most clinicians would probably use more focused tests to address specific referral questions rather than administer a test such as this.


References

Cizek, G. J., & Sandoval, J. (2003). Review of the Woodcock-Johnson III. In B. S. Plake, J. C. Impara, & R. A. Spies (Eds.), The fifteenth mental measurements yearbook (pp. 1019-1028). Lincoln, NE: Buros Institute of Mental Measurements.

Mather, N., & Woodcock, R. W. (2001). Woodcock-Johnson III tests of achievement examiner's manual. Itasca, IL: The Riverside Publishing Company.

McGrew, K. S., & Woodcock, R. W. (2001). Woodcock-Johnson III technical manual. Itasca, IL: The Riverside Publishing Company.

Woodcock, R. W., McGrew, K. S., & Mather, N. (2001). Woodcock-Johnson tests of achievement III (WJ-III). Rolling Meadows, IL: Riverside Publishing.

To cite this document:

Hayward, D. V., Stewart, G. E., Phillips, L. M., Norris, S. P., & Lovell, M. A. (2008). At-a-glance test review: Woodcock-Johnson tests of achievement III (WJ-III). Language, Phonological Awareness, and Reading Test Directory (pp. 1-4). Edmonton, AB: Canadian Centre for Research on Literacy. Retrieved [insert date] from http://www.uofaweb.ualberta.ca/elementaryed/ccrl.cfm.