Public Finance Proposal Draft

Alex Fernandez

Public Finance

October 3, 2012

Public Finance Proposal Draft

Class size and the effect of smaller class sizes have been researched extensively throughout the economic literature. Class size reduction (CSR) has been suggested as a policy solution to enhance educational achievement in struggling schools.This paper aims to address the literature on class size in order to understand how large the effect of class size actually is on scholastic achievement.Furthermore, this paper seeks to find who is benefited the most by a class-size reduction policy,and for which subject areas. The Tennessee STAR project is often cited as the pinnacle of class-size experiments that have been conducted. The STAR project and a number of non-experimental research will be evaluated.

Note: At this point I have been reading a number of papers but have yet to develop a comprehensive report on my findings. Below are a number of quick points from the articles I have read thus far.

Literature Review:

“Some Findings from an Independent Investigation of the Tennessee STAR Experiment andfrom Other Investigations of Class Size Effects”

Eric A. Hanushek

Educational Evaluation and Policy Analysis, Vol. 21, No. 2, Special Issue: Class Size:Issues and New Findings (Summer, 1999), pp. 143-163

Hanushek investigates non-experimental results of historical class-size adjustments. He suggests that the results from the Project STAR experiment need to be viewed with consideration of possible implementation and randomness errors. He also points to problematic results from a possibly biased attrition in the sample. Hanushek argues that there is considerable uncertainty in the validity of Project STAR results.

“Do Low-Achieving Students Benefit More from Small Classes? Evidence from the Tennessee Class Size Experiment.”

Barbara Nye, Larry V. Hedges and Spyros Konstantopoulos

Educational Evaluation and Policy Analysis, Vol. 24, No. 3 (Autumn, 2002), pp. 201-217

This article studies the data from the Tennessee STAR experiment and finds that there are not statistically significant differential effects for class size reduction on low-achieving students when compared to high-achieving students. Further, this study finds that for mathematics the effect of small class size on low-achievers is smaller than that of high-achievers. Though there appears to be an overall positive effect of smaller class size for these students, there does not appear to be a differentially greater effect on low-achieving students.

Answers and Questions about Class Size: A Statewide Experiment

Jeremy D. Finn and Charles M. Achilles

American Educational Research Journal, Vol. 27, No. 3 (Autumn, 1990), pp. 557-577

In this article the authors use a large-scale longitudinal study to test the effect of class size reduction on reading and mathematics following a group of students over a two-year span. The authors find that class size reduction has a significant positive effect on results for standardized and curriculum based tests. In this study the authors also found that minority students were particularly benefitted by the smaller class size. The smaller classes in the study outperformed the larger classes and appeared to acquire greater gains in reading outcomes.

The Effects of Class Size on Student Achievement: New Evidence from Population Variation

CM Hoxby

The Quarterly Journal of Economics, 2000

In this article the author uses two types of models which both result in little or no effect on achievement with class size reduction. The author posits that since teachers do not always teach smaller classes they do not utilize specialized curriculum and do not understand how to take advantage of a smaller class size. The author finds no significant effect in both types of models she uses and additionally she finds no evidence that class reductions have a greater effect on low-income or minority students.

Economic Considerations and Class Size

Krueger, Alan.

The Economic Jounal, Vol. 113, No. 485, pp. F34-F63.

In the paper Krueger investigates evidence on the effect ofclass sizeon student achievement. He notes that Hanushek’s classification of estimates and weighting scheme in analyzing class size is flawed and leads to inconclusive results.A cost-benefit analysis of class size reduction is performed and the results of the Tennessee STAR class-size experiment suggest that the internal rate of return from reducing class size from 22 to 15 students is around 6%.