Spelling Subtests Research Articles

Concurrent Validity of the Spelling Subtest of the Peabody Individual Achievement Test Relative to Dictated Spelling Tests: An Experimental Analysis of Selection Versus Supply as Response Demands

The purpose of the study was to investigate the concurrent validity of the spelling subtest of the Peabody Individual Achievement Test (PIAT) relative to the common classroom evaluation method of dictated spelling tests among learning disabled students. Subjects were 15 learning disabled children ranging in age from 6 years 10 months to 9 years 10 months. The spelling subtest was administered twice to each subject. In the PIAT-Regular condition, the subtest was administered according to instructions in the PIAT manual. The PIAT-Dictated condition conformed to instructions in the manual with the exception that the subjects were asked to spell dictated test items (supply) rather than choose the correct spelling from among four choices (selection). A dependent t-test indicated that subjects performed significantly better under the PIAT-Regular condition than under the PIAT-Dictated condition. A Pearson product-moment correlation coefficient indicated that the rank order of the subtest scores did not change substantially from one condition to the other. Implications of the results were discussed.

The Concurrent Validity of the Woodcock Reading Mastery Tests Relative to the Peabody Individual Achievement Test among Retarded Adolescents

To investigate the concurrent validity of the Woodcock Reading Mastery Tests (WRMT) and the Peabody Individual Achievement Test (PIAT), both instruments were administered to 56 adolescents. Scores on the PIAT reading and spelling subtests were correlated with scores on the WRMT subtests which were designed to measure similar skills. Partial correlation coefficients were calculated for male, female, black, and white children as well as for the total group. The presence of relatively high and positive partial correlations between appropriate subtests supported the concurrent validity of the WRMT relative to the PIAT and to criterion variables. Grade equivalents were higher on the Peabody than on the Woodcock. Possible interpretations of the results were discussed.

The WISC-R as a Predictor of WRAT Achievement for Educable Mentally Handicapped Students

The present study investigated the predictive validity of the Wechsler Intelligence Scale for Children-Revised (WISC-R) Full Scale IQ relative to the Wide Range Achievement Test (WRAT) Reading, Spelling, and Arithmetic subtests as criterion measures for a group of 61 educable mentally handicapped (EMH) students in a public school setting. The results of a multivariate multiple regression analysis indicated that WISC-R Full Scale IQ did not yield valid prediction of achievement as measured by the WRAT subtests. Cautions are presented with regard to generalizing the current findings to other student populations.

Childhood cancer and vulnerability for significant academic underachievement.

Learning difficulties related to childhood cancer with its intense physical and psychosocial stresses are examined by comparison of 22 oncology patients and 22 children referred for psychiatric/psychological evaluation. Findings demonstrate, among children undergoing treatment of cancer, some academic vulnerabilities for which psychosocial aspects may not fully account and which become prominent under comparisons of within-subject differences between normalized scores of the WISC-R Verbal IQ and the PIAT Reading (p < .05) or Spelling (p < .01) subtests. Moreover, when members of the oncology group, who had already been placed in learning disabilities programs, were added to those defined as significantly underachieving by McLeod's method, it appeared that nearly a third of the oncology sample were encountering marked academic inhibitions.

Comparison of 1965, 1976, and 1978 Norms for the Wide Range Achievement Test

A comparison of the 1965 and 1978 Wide Range Achievement Test norms indicates that the 1978 norms, generally, with the exception of the youngest age levels, provide lower standard scores on the reading and spelling subtests. Arithmetic norms, however, show a more variable pattern of change, with the 1978 norms frequently providing higher standard scores than the 1965 norms for the same raw scores. Most of the variation between the two sets of norms occurs primarily in the upper or lower portions of the ability range. Few changes appear between the 1976 and 1978 norms. Grade ratings for raw scores are the same for 1965 and 1976, but they are changed for 1978, with the 1978 grade ratings being markedly lower for the highest raw scores. Changes may be related to changes in ability level or to changes in the normative sample.

Wide Range Achievement Test: A One-Year Stability Study

This study examined the stability of the 1965 and 1978 Wide Range Achievement Test standard scores for a total sample of 115 kindergarten, first, and second grade students on the Hopi Indian Reservation in Northern Arizona. Results indicate satisfactory stability over a 1-yr. interval for the Reading, Spelling, and Arithmetic subtests using both the 1965 and 1978 editions. At the same test administration, the 1965 and 1978 standard scores correlated significantly with each other. Standard score subtest means derived from the 1965 and 1978 editions of the WRAT were compared and determined to be, in most cases, significantly different from one another.

The Concurrent Validity of the Larsen-Hammill Test of Written Spelling in Relation to the California Achievement Test

The Larsen-Hammill Test of Written Spelling (TWS) scores were correlated with scores received on the California Achievement Test (CAT) by 45 fourth-grade pupils. The two subtest scores on the TWS were used as predictor variables in four multiple regression analyses in which combined Reading, Mathematics, and Language scores along with the Spelling score from the CAT were used as criterion variables. The linear combination of both the TWS predictor variables, Predictable Words and Unpredictable Words, most accurately forecast the CAT Language scores. CAT Spelling scores were significantly predicted by TWS Predictable Words scores, but not so accurately predicted by the addition of the TWS Unpredictable Words scores into the regression equation. The investigation did not lend support to the contention that the TWS yields the same results as the CAT Spelling subtest or other subtests of the CAT.

Learning-Disabled Boys as Adolescents: Cognitive Factors and Achievement

Ninety-three boys, classified in grade school as either learning-disabled (LD) or normal achievers (controls), were followed-up at age 14. The 62 LD returnees remained, as a group, seriously retarded on the Gray Oral Reading Test and on the spelling and arithmetic subtests of the Wide Range Achievement Test (WRAT), but were just over a year behind (age norms) on the WRAT reading test. Controls averaged 10th-grade level on both reading tests and at grade level on the WRAT spelling and arithmetic tests. LD returnees wrote significantly slower than controls, their own names as well as a word and short sentence. Students who had poorest outcomes at follow-up were those who had the most severe early reading problems and who had depressed scores on the Information, Arithmetic, Digit Span, and Coding subtests of the WISC. This cluster continued to separate LD students and controls at age 14. Achievement-level outcomes were not related to earlier classification by activity level and neurological maturity on to intervention efforts.

Concurrent Validity of the Peabody Individual Achievement Test

The concurrent validity of the PIAT was investigated using the WRAT and WISC. Pearsonian correlations between the PIAT, WRAT, and selected WISC subtests, along with means and standard deviations for the WISC are presented for 40 children comprising Normal, Reading Disabled, and Mentally Retarded groups. Higher correlations were found between the reading subtests while only moderate correlations were found for the spelling and mathematics subtests. It was concluded that the PIAT may be used with the WRAT and WISC to facilitate differential diagnosis.

RELATIONSHIP OF ELEVATED BLOOD TYROSINE TO THE ULTIMATE INTELLECTUAL PERFORMANCE OF PREMATURE INFANTS

Blood tyrosine concentrations were followed from birth to nursery discharge in 71 premature infants fed a high protein formula supplemented by 60 mg/day of ascorbic acid. In 89% of infants blood tyrosine concentrations were abnormal, and in 38% of infants the maximum level observed was 15.0 mg/100 ml or higher. Maximum blood tyrosine levels correlated significantly with gestational age (p = < 0.05) but not with birth weight. In a follow-up study performed at 15 months of age, infants with high tyrosine levels had no increase in the incidence of neurological abnormalities. Between 7 and 8 years of age a second follow-up study was performed on 62 children. This included a WISC, a Wide-Range Achievement Test (WRAT), and tests for psychomotor and language maturity. Two children had died in the interval, and five of the 62 were retarded for full testing. The full scale WISC I.Q. of all children correlated with birth weight at the 10% confidence level (p = < 0.1). The mean WISC I.Q. of high and low tyrosine subjects was 82.9 and 81.6 respectively. When infants were grouped by birth weight, a significant difference was detected in subjects weighing 2,000 gm or more. High tyrosine infants had a significantly lower performance I.Q. than low tyrosine infants (82.4 and 97.8 respectively; p = < 0.02). Significant differences were recorded in the scores on Object Assembly, Picture Assembly, and Picture Completion of the WISC. Significant differences were also seen on the Spelling subtest of the WRAT (p = < 0.02). We observed no adverse effect of high tyrosine levels on the intellectual performance of smaller premature infants, who on the whole have a greater risk for other complications of prematurity.