Psychology and Neuroscience Faculty Database
Psychology and Neuroscience
Arts & Sciences
Duke University

 HOME > Arts & Sciences > pn > Faculty    Search Help Login pdf version printable version 

Publications [#42169] of Jeffrey C Valentine

search PubMed.

Peer-reviewed journal articles published

  1. Cooper, H., Valentine, J.C., Charlton, K., & Melson, A. (2003). The effects of modified school calendars on student achievement and school community attitudes: A research synthesis. Review of Educatonal Research, 73, 1-52.
    (last updated on 2005/10/13)

    Abstract:
    Many proponents of school calendar change call for modified arrangements in which children might or might not attend school for more days but the long summer vacation is replaced by shorter cycles of attendance breaks. For example, children might go to school for 9 weeks and then get 3 weeks vacation, or go 12 weeks and then get 4 weeks vacation. This review focuses on studies of school districts that modified their calendars but did not increase the length of their school year. We briefly describe (a) some background material on school calendar issues, (b) the arguments put forth by both the proponents and opponents of modified school calendars, and (c) previous reviews of the literature on the effect of modified calendars on achievement test scores. Then, we present a research synthesis that attempted to retrieve the largest possible body of empirical evidence on modified school calendars, including sources available from both the proponents and opponents of calendar change. Effect size estimates were combined using conservative assumptions regarding the independence of effect estimates and estimates of error. A statistical synthesis was conducted not only on student achievement but also on data relating to the attitudes of students, parents, teachers, administrators, staff, and the general public who had experienced modified calendars. The most important finding of the synthesis was that the quality of evidence available on modified school calendars leaves much to be desired. Most evaluations made no attempt to equate groups of students who attended modified calendar schools and their traditional calendar counterparts, beyond choosing a similar school in the same school district. Even within this weak inferential framework, the evidence from the meta-analysis revealed ambiguous results. First, a vote-count revealed that 62% of 58 districts reported that students in the modified calendar program outperformed students in the traditional calendar program. Second, the average weighted effect sizes for 39 school districts was d=.06, favoring modified calendars. Third, 17 school districts that attempted some statistical or matching control revealed an average weighted effect size of d=.01 using a fixed model of error, and d=.11 using a random error model. This higher estimate was significantly different from zero. There was evidence that modified calendar programs do improve achievement for economically disadvantaged or poor-achieving students, that programs implemented more recently may be showing improved results, and that the students, parents, and staffs who participate in modified calendar programs are overwhelmingly positive about the experience. There are also specific actions that policy makers can take to improve community acceptance of modified calendars, such as involving the community in planning the program and providing quality intersession activities.


Duke University * Arts & Sciences * Faculty * Staff * Grad * Postdocs * Reload * Login