Growth in Standardized Ability and Achievement Test Scores From 3rd to 10th Grade.

Abstract

Growth in Test of Cognitive Skills (TCS) scores and Comprehensive Tests of Basic Skills (CTBS) reading, math, and total achievement scores from 3rd to 10th grade was studied in 328 public school students in a middle-class suburban community. Surprisingly, groups differing in ability and achievement in 3rd grade made parallel progress over time, and some “fan-close” effects were found. With growth curve analysis of individual students using hierarchical linear models, initial status on cognitive ability predicted initial achievement scores but did not affect the rate of growth. Similarly, initial status in achievement predicted the intercept but not the slope in cognitive ability scores over time. Although replication is needed, this study illustrates how districts could use standardized test data to document growth of academic skills over time in high-, middle-, and low-achieving children; in racial or socioeconomic status subgroups; or in different classrooms, schools, or districts. On January 8, 2002, President Bush signed into law the No Child Left Behind Act of 2001, which dictates that any state obtaining federal funds must show that it has adopted a program of high-quality, annual student assessments that can be used to determine the annual performance of every school and local educational agency (No Child Left Behind Act, 2002). Although the rhetoric surrounding passage of this bill suggested that regular standardized testing is a novel idea, in fact most school districts have well-established testing programs using both commercial tests and state-based assessments. Surprisingly, however, there is little research on such test data. In the present study, we analyzed standardized test data from one school district for three cohorts of children, each tested in 3rd, 5th, 8th, and 10th grades. Repeated measures analysis of variance (ANOVA) and hierarchical linear modeling (HLM) were used to analyze performance over time as a function of initial status on ability and achievement measures. Particularly because the results of our study were surprising, replication in other school districts should be a high priority for future research.