Achievement Prediction of ‘High-Risk’ Children

Abstract

Of 300 kindergarten children in a suburban school system, 62 were identified as possible learning problems on the basis of teachers' ratings and comprehensive test batteries. In considering the different scales in terms of their ability to predict first grade achievement for these children, the best predictors were (a) teachers' ratings, (b) the various number pretests (e.g., Digit Span, Arithmetic), (c) the reading readiness rating derived from the Metropolitan Readiness Tests, and (d) knowledge of letter names. Although many of the tests showed poor predictive value for ‘high risk’ children, this is in no way a comment on their usefulness as diagnostic or clinical instruments for other purposes.

Children Who were Very Low Birthweight: Cognitive Abilities and Classroom Behavior at Five Years of Age

“…This task-oriented behavior, which was rated in May at the end of the school year, when children should have adjusted to teacher expectations for task performance, has been noted by Simmer (1983) as an "effective early warning" sign for identifying children in kindergarten who are at risk for later learning problems. Furthermore, Serwer, Shapiro, and Shapiro (1972), as well as studies noted earlier, found significant correlations between teacher-rated task-oriented behavior and achievement during the elementary years. Meyers, Atwell, and Orpet (1968) found that the behavioral rating "attention" in kindergarten emerged as the most powerful predictor for three of the subtests of the California Test of Achievement in fifth grade.…”

Measuring intelligence with the Goodenough-Harris Drawing Test.

“…Strumpfer and Mienie (1968) obtained a split-half reliability of .88 for the Man scale and .92 for the Woman scale by the Spearman-Brown method. Serwer, Shapiro, and Shapiro (1972) reported a similar split-half reliability of .87. Sinha (1970) obtained interitem reliability coefficients between .84 and .90 by the Kuder-Richardson Formula 20.…”

“…For the most part, the GH accounted for less than 10% of the variance in measures of academic achievement, raising the question of the test's theoretical soundness as an intellectual measure as well as its practical utility as a screening device. The incremental validity of the GH in relation to other predictors was assessed in six studies (Dudek et al, 1969 1967c; Eaves, Kendall, & Critchton, 1972;Hall & Chansky, 1971;Henderson et al, 1973;Panther, 1967;Pihl & Nimrod, 1976;Serwer et al, 1972;Stevenson, Parker, Wilkinson, Hegion, & Fish, 1976). In each case, the GH added so little to the battery's predictive ability that it was excluded from the final regression equation.…”

The magical numbers 7 and 4 are resistant to the Flynn effect: No evidence for increases in forward or backward recall across 85 years of data

“…Digit Span has also been found to correlate with years of education completed (r = .44, Paul et al, 2005;r = .43, Birren & Morrison, 1961), reading comprehension (r = .30; Daneman & Merikle, 1996;Norman, Kemper, & Kynette, 1992), and brain volume (r = .41; Wickett, Vernon, & Lee, 2000). Additionally, amongst a battery of cognitive ability tests, Digit Span was found to be the best predictor of academic achievement amongst learningproblem children (Serwer, Shapiro, & Shapiro, 1972). Digit Span has also been found to be a respectable predictor of job performance (medium cognitive demands: r = .51; Verive & McDaniel, 1996).…”