Impact of experimental design on effect size: Findings from the research literature on training.

This article examines whether differences in the equations commonly used to calculate effect size for single group pretest-posttest (SGPP) designs versus those for control group designs can account for the finding that SGPP designs yield larger mean effect sizes (e.g., M. S. Lipsey and posttest only with control, POWC). Across 45 meta-analyses , mean effect sizes (dvalues) for SGPP designs were 0.29 SDs larger (0.76 versus 0.47) than effect sizes for studies using control or comparison group designs. Research by Guzzo, Jette, and Katzell (1985) has found similar differences. The common conclusion among these researchers is that data from SGPP designs clearly are capable of upwardly biasing the mean treatment effect estimates derived from meta-analysis (Lipsey & Wilson, 1993, p. 1194). Lipsey and Wilson's (1993) solution to the perceived upward bias in effect size for SGPP designs was not to