An application of hierarchical linear growth modeling: An examination of growth trajectories of subgroups not meeting adequate yearly progress in mathematics

Illinois utilizes a status model to evaluate districts' adequate yearly progress (AYP) in compliance with the No Child Left Behind legislation. This study applied growth modeling to determine if significant growth occurred despite a district's failure to meet AYP. The study examined the mathematics achievement of two cohort groups of students.;The students were enrolled in a district outside of Chicago. Of the 960 students in Cohort 1, 468 were female and 492 were male. Seventy-five percent of the students were classified as non low-income. The number of White students represented 62% of the population, the number of African-American students represented 22% of the population, and the number of Hispanic students represented 16% of the population. Of the 973 students in Cohort 2, 502 were female and 471 were male. Sixty-seven percent of the students were classified as non low-income.;The number of White students represented 57% of the population, the number of African-American students represented 27% of the population, and the number of Hispanic students represented 16% of the population.;For each cohort, the students' 8th grade EXPLORE score, the students' 9th and 10th grade PLAN score, and the students' 11th grade ACT score were collected. The study was designed to answer the following research questions: (1) Is there significant variability across students' mathematics achievement at 10th grade? (2) What is the average rate of growth in mathematics achievement at 10th grade? (3) Is there acceleration or deceleration to the mathematics growth from 8th through 11th grade? and (4) Are there any time-invariant predictors such as gender, low-income status, building or ethnicity that could explain between-student differences in mathematics achievement?;The results indicate significant variability across students' predicted average mathematics achievement at 10th grade. For both cohorts, the quadratic growth parameter was significant and indicated acceleration in predicted math achievement from 8th to 11th grade. The most striking finding in the study was that for the African-American students in both cohorts, their predicted math achievement and their predicted instantaneous rate of growth were below White students, and there existed deceleration in their predicted rates of growth over time.