Although a number of research papers conclude that grouping students by ability – called "tracking," "ability grouping" or "achievement grouping" – is harmful, especially to those placed on low-ability groups, a recent study from the National Bureau of Economic Research disputes those finding. NBER research seems to indicate that sorting by achievement actually has a positive impact on academic outcomes of both low-achieving and high-achieving students both.
Yet a new Think Twice publication taking on the NBER report argues that a number of methodological errors were made in its composition, thus putting its conclusions into question.
Dr. Carol Corbett Burris, principal of South Side High School in the Rockville Centre School District in New York, and Katherine Allison, a doctoral student in research and methodology at the University of Colorado Boulder, reviewed the report, Does Sorting Students Improve Scores? An Analysis of Class Composition, for the Think Twice think tank review project. Burris is the co-author of two books on tracking and equity as well as numerous articles regarding these issues in peer-reviewed and popular journals. The review was prepared for the National Education Policy Center with funding from the Great Lakes Center for Education Research and Practice.
For its dataset, the original paper uses standardized test scores from exams administered in Texas and concludes based on score comparisons between tests given in 3rd and 4th grade that sorting by ability is associated with substantial academic gains. But those gains don't show up when looking at exam results by sub-group, and relies on the assumption that score distributions over the entire population of test takers replicates over all demographic and ability groups.
Furthermore, Think Twice says the report doesn't examine or provide details about how students were sorted in the first place and how schools and districts went about determining into which group each student falls.
The report also depends on relative performance of groups of students on two proficiency tests rather than looking for indication of academic growth specifically. All these issues drive Think Twice's conclusion that the original paper isn't robust enough to "inform policy."
Unlike the majority of studies of tracking, which identify sorting practices through school reports, this study seeks to identify schools that sort students for instruction through an examination of the spread of scores in any given classroom. It ignores the variety of ways that elementary schools group for instruction,2 however, such as within-class grouping, flexible grouping, re-grouping for specific subjects and between-class grouping. Principals of elementary schools may also assign particular students to particular teachers based on past effectiveness with those or similar students.