Richard L. Wikoff

The WISC-R as a Predictor of Achievement.

This study investigated the validity of the WISC-R IQs as predictors of achievement as measured by the PIAT. Both tests were administered to 188 children referred to the author because of learning problems. Regression analyses were run with each of the PIAT subtests and the total score as criteria and the WISC-R IQs as predictors. Each of the IQs was a significant predictor of each of the PIAT subtests, but the Verbal IQ predicted best. Regression equations were provided and several indications of the importance of the results were presented.

Factor analysis of four Fennema‐Sherman mathematics attitude scales

The factorial structure and reliability of four Fennema‐Sherman mathematics attitude scales were investigated. The scales were administered to 204 college students in five mathematics classes and submitted to factor analysis. The scales were found to be reliable but measured three factors instead of four. The scales appeared to be generalizable to college populations, although the items may need to be reorganized. Additional research is needed to determine their usefulness for counselling or for prediction of success in math classes. Future research should consider the possibility of sex differences, also.

Prediction of the WISC-R full scale IQ from the SFTAA

Investigated the validity of the Short Form Test of Academic Aptitude as a predictor of WISC-R Full Scale IQ. A multiple regression equation was computed for 126 Ss, who had been administered both tests using the WISC-R Full Scale IQ as the criterion variable and the Language and Nonlanguage scores of the SFTAA as predictor variables. The multiple correlation was .73, which indicated satisfactory validity. The regression coefficients are presented.

Effectiveness of a personalized system of instruction in an algebra/trig sequence at the college level

A pre‐test/post‐test experimental model was used to study the effectiveness of a personalized system of instruction (PSI) in the algebra/trig sequence at the college level. Data were obtained from 777 students who completed a course in algebra, pre‐calculus algebra, or trigonometry. Students were divided into control groups taught by the lecture‐discussion method with one instructor per section and experimental groups taught by a modified Keller PSI plan. Both groups used the same text and syllabus. Using an analysis of covariance controlling for GPA, the mean gain score of the experimental group was significantly larger than the mean gain score of the control group for algebra (p = 0.056, N=503). The results were not significant for the other two classes.

Subscale classification schemata for the Stanford-Binet, form L-M

366 Stanford-Binet (S-B) protocols were scored according to subscale categories proposed by Valett (1963–64), Sattler (1965), and Meeker (1969). The mean age of the Ss was 9.24 years and the mean IQ was 99.43. Intercorrelations were computed for the subscales, sex, socioeconomic status, mental age, and vocabulary score and factor analyses performed on each of the subscale intercorrelation matrices. The intercorrelations were significant for all variables except sex. Only one general factor was found for each of the schemata. It was concluded that in the absence of factorial validity, the use of such subscales is invalid.

Home Environment and Adjustment in Families with Handicapped Children: A Canonical Correlation Study

The hypothesis that adjustment within the family of a handicapped child is due to more than just the presence of the child was investigated using canonical correlation. Five borne environment variables—handicapped-related events, resources available, severity of symptoms, other sources of stress, and socioeconomic status—and three adjustment variables—subjective assessment of the childs presence, marital adjustment, and maternal health—were used. Data were collected from 105 mothers of handicapped children. Two significant variates emerged with canonical correlations of .776 and .447. The total redundancy for the adjustment variables across both variates was .346. For the environment variables, the total redundancy was .251. The resources available and the number of handicapped-related events were the most important environment variables, and socioeconomic status was the least important.

Validation of the M.A.A. tests as predictors of success in algebra/trig sequence of courses

Three tests published by the Mathematics Association of America were administered during the first week of classes to students enrolling in algebra, pre‐calculus algebra, and trigonometry classes. Scores were correlated with grades using eta and Pearson correlations. Eta coefficients were significant for all classes. The Pearson correlations were significant for the two algebra courses but not for the trigonometry course. It was concluded that the algebra tests were valid as predictors of course grade but no support was found for the validity of the trigonometry test.

Determining basals for the peabody individual achievement test

The appropriateness of the starting points for PIAT subtests as suggested by the test authors was investigated. For a sample of 113 subjects with an average WISC-R Full Scale IQ of 98, there was an average difference of nearly 8 points between the suggested starting points for the Mathematics subtest and the actual basals. The difference between suggested starting points and basals for the other subtests ranged from 8 to 12 points. It was suggested that the PIAT user start each subtest approximately 8 items below that suggested by the Manual.