Title
sizefx -- Calculate Cohen's d, Hedges' g, and the effect size (ES) correlation > (r)
Syntax
sizefx varlist [if] [in]
Description
sizefx computes Cohen's d, Hedges' g, and effect size (ES) correlations (r) for the two variables specified in varlist. The two variables specified correspond to the variables in a ttest, although the "==" operator is not used in the sizefx syntax.
The command displays Cohen's d statistic (Cohen 1988) using the pooled variance of the variables and Hedges' g statistic. The pooled variance for an unpaired t-test is used to calculate effect size even when using a dependent sample or correlated designs (Dunlap et al. 1996).
Additionally, it displays the effect size (ES) correlation (r) using the value of Cohen's d.
Examples
. sizefx age1 age2 . sizefx before after
References
Cohen, J. (1988) Statistical power analysis for the behavioral sciences, 2nd ed. Hillsdale, NJ: Lawrence Earlbaum Associates.
Dunlap, William P., Cortina, Jose M., Vaslow, Joel B., and Burke, Michael J. (1996) Meta-analysis of experiments with matched groups or repeated measures designs. Psychological Methods 1 no. 2, 170-177.
Thalheimer, W., and Cook, S. (2002, August) How to calculate effect sizes from published research articles: A simplified methodology. Retrieved 01 Apr 2010 from http://work-learning.com/effect_sizes.htm.
Additional resources may be found at: http://www.gseis.ucla.edu/courses/ed230a2/notes/effect.html http://web.uccs.edu/lbecker/Psy590/es.htm
Links active as of 01 Apr 2010.
Author
Matthew S. Openshaw, The University of Texas at Dallas, USA openshaw@utdallas.edu
Please send comments, suggestions or constructive criticism to the above email address.
sizefx v1.3 Update 01 Apr 2010