Monday, March 7, 2016

Exhaustive Reviews of Research on Single-Sex Education

A brief quote from an article of Dr. Cornelius Riordan titled Schools, Single-Sex in the Blackwell Encyclopedia of Sociology 2015.

Dr. Riordan says that there are three exhaustive reviews of research on single-sex schools so far.

First, the research of Moore, Piper, and Schaefer (1992) for a US Department of Education report. This found evidence that "clearly supported the proposition that single-sex schools may produce positive outcomes, especially for young women."

Second, the study of Mael et al (2005) entitled Single-sex versus coeducational schooling: A systematic review, also for the U.S. Department of Education. This used What Works Clearinghouse standards. The review, according to Riordan, found that "studies preponderantly support the view that single-sex schooling has positive benefits for both sexes, in terms of both academic short term achievement and socioemotional development."

Lastly, Riordan refers to the meta-analysis of Pahlke, Hyde, and Allison (2014) which concluded that “there is little evidence of an advantage of SS schooling for girls or boys for any of the outcomes.”

Riordan says that "It is important to note that the authors employ a 0.2 effect-size threshold in drawing these conclusions about there being no advantage to single-sex schooling. Despite the above conclusion, the research found that, in a separate analysis of just the best studies (well controlled) conducted in America, the effect size in mathematics was 0.14 for both boys and girls. The verbal performance was 0.22 for girls and 0.13 for boys.

These math and verbal outcomes in favor of single-sex schools parallel the findings in Mael et al. (2005). Sound educational research has shown that a standard effect size of 0.10 on gains from sophomore to senior year of high school is equivalent to one full year of learning by the average public school student in the United States (Hoffer, Greeley, and Coleman, 1985). Applying this standard, a difference of 0.10 (or greater) between students in single-sex and in coeducational schools would be substantially important, whether as a gain score or as a cross-sectional effect, assuming that it was statistically significant."

No comments: