A recent study of kindergarten classes revealed significantly different results than previous research. Jeremy D. Finn and Gina M. Pannozzo, State University of New York/Buffalo, examined the conditions that promote or discourage engagement in kindergarten classrooms and achievement at the end of the year.
Using data on over 10,000 kindergarten students from a national sample of kindergarten students in the Early Childhood Longitudinal Study by the U. S. Department of Education, they compared the academic engagement of children in large and small classes, in full-day and half-day programs, and in classes that had teacher aides. Engagement referred to both learning and social behaviors. Teachers rated a range of behaviors for both individual students and for whole classes.
The strongest correlations with spring achievement tests were found for behavioral approaches to learning by individual students. Basic behaviors such as attentiveness and task persistence which a child needs in order to learn from classroom instruction were the most strongly related to achievement. Social behaviors were not as strongly correlated with math and reading scores.
Students fared better in half-day programs
Finn and Pannozzo studied the features of classroom organization–presence of an aide, length of day and size of class–to determine their effect on student behavior. Length of day and class size were significantly related to overall class behavior, with smaller, half-day classes showing more appropriate and productive behavior. This is in contrast to previous research that revealed the benefits of full-day kindergarten. However, teachers’ ratings of individual students were not significantly correlated with either size of class or length of day. And the presence of an aide had no positive effects on performance and behavior at either the individual or class level.
These researchers call for further research to clarify the disparity between class- and student-level results. They suggest that this disparity may reflect sampling issues: for example, the survey may have had too few students per class to give reliable estimates of class averages. It is also possible that there might be methodological differences between individual and class ratings that affected results. Finn and Pannozzo recommend that observations by outside professionals as well as classroom teachers be used to measure behavior.
“Classroom Organization and Student Behavior in Kindergarten”, Journal of Educational Research, Volume 98, Number 2, December 2004, pp. 79-91.
Published in ERN January 2005 Volume 18 Number 1