Woodcock Reading Mastery Tests Revised

The Woodcock Reading Mastery Tests Revised (WRMTR; published by American Guidance Service, http://www.agsnet.com/) are a set of scales that take 40–50 minutes to administer en toto and that assess reading achievement. The WRMTR contains tests of visual-auditory learning and letter identification, which comprise the Reading Readiness Cluster. The word identification and word attack tests combine to form the Basic Skills Cluster, and the word comprehension and passage comprehension tests make up the Reading Comprehension Cluster. The word comprehension test comprises three subtests: synonyms, antonyms, and analogies.

The WRMTR has two forms, only one of which (i.e., Form G) contains the Reading Readiness Cluster tests. The tests are intended for use with individuals from kindergarten through college-age students and with older adults (75 years and older). With the exception of the visual-auditory learning and word attack tests, starting point, basal, and ceiling rules are consistent throughout the battery. The two exceptions cause confusion for some examiners. Hand scoring can be time consuming and prone to error, but the available scoring software reduces scoring time by 70 to 80 percent. The WRMTR provides a full range of score types: age and grade equivalents, standard scores, percentile ranks, relative performance indices, and instructional ranges (i.e., easy, instructional level, and difficult). Standard errors of measurement allow the hand scorer to construct 68% confidence intervals to estimate the true score. Standard errors of differences, which would allow the user to estimate an examinee's strengths and weaknesses, are not available in the manual or the scoring software. Instead, the test author recommends profile analysis, which has inherent weaknesses and provides information that is prone to error.

New norms, which include only grades kindergarten through high school, were collected in 1995 to 1996 and were based on 1994 census data. Approximately 3,400 individuals were tested, though the exact number varies from one test or cluster to another. The sample was controlled for age, gender, race, region, socioeconomic status, community size, and special populations. Nevertheless, Native Americans, Asians, and Pacific Islanders are overrepresented, and the northeastern states are underrepresented in the WRMTR sample.

The author reports only split-half reliability coefficients. The median internal consistency reliability coefficient for the tests is .91 (range = .68 to .98). For the clusters, the median reliability coefficient is .95 (range = .87 to .98). For the full-scale score, the median reliability coefficient is .97 (range = .86 to .99). The lack of alternate-form and test-retest reliability estimates is a serious omission for this highly popular test battery.

Very little information on the validity of the WRMTR is presented in the manual; much of it is based on the original (1973) version of the battery. Presumably, independent research will clarify the validity of this set of instruments, but it remains incumbent on the author of the assessment tool to provide evidence of the technical adequacy of his product.