""""""""Mapping"""""""" Tracking in the Elementary Grades: A Nationwide Longitudinal Inquiry. In this project we conduct the first comprehensive, longitudinal examination of ability grouping in the early elementary grades in the United States, with specific attention to within-class ability grouping and its impact on children's cognitive growth. Our approach has both theoretical and practical policy implications because it provides needed knowledge about how homogeneous ability grouping structures learning opportunities for children pertaining to their reading achievement. The recent release of the Early Childhood Longitudinal Study-Kindergarten Class of 1998-99 (ECLS-K) data, conducted by the National Center for Educational Statistics, provides an unprecedented occasion to study this issue. Despite a vast extant literature pertinent to this subject, previous studies do not have many of the features we propose that are necessary to truly understand homogeneous ability grouping and its effects. ? ? We plan to conduct this project in two phases. In Phase I we attempt to sort out the forms of ability grouping by focusing on within-class ability grouping, the most extensive form of tracking in these grades. We examine the types of schools and classrooms in which within-class ability grouping is used and the types of students who get placed in different ability groups. In Phase II, we utilize research findings of the first phase to construct multivariate longitudinal models that estimate the cognitive growth of young children. The extent to which ability grouping is practiced ranges considerably from one classroom to another, therefore, the extensive information on ability grouping from kindergarten through the third grade made available through the ECLS-K can provide new insights concerning the impact of ability grouping. In the second phase, we also test whether changes in measurable performance are reflected in mobility from low to high ability groups. Unlike prior studies, we estimate the effects of ability grouping by comparing students with similar cognitive attributes who are in grouped and ungrouped classrooms. We employ Hierarchical Linear Modeling and conduct stratified propensity analysis to correct for the possibility of selection bias in students' ability group placements. This research allows us to consider the theoretical and policy implications of social reproduction versus meritocratic perspectives with respect to the impact of segregating children for instruction on the basis of their measured cognitive performance. ? ?
