I had a question about how to calculate Cohen's D a particular situation for my metaanalysis. I am a first year grad student, so my familiarity with statistics is somewhat limited!
The study design  participants are assigned to one of two groups (parallel design), but not a true RCT because both are interventional (no control group).
What I want to calculate  I want to include both interventions in my metaanalysis, and calculate an effect size for each of them (cohen's D). I am thinking I will compare the pre and post intervention scores for each group, for my effect sizes.
Information I have  I have pre and post means and standard deviations for each of the groups for my outcome measure. I also have the sample size of each group, of course. The paper, also reports the effect sizes (using ηp^2) for a 2 X 2 ANOVA (intervention groups  intervention1, intervention 2  BY time  baseline, post)...for the "time factor" of one of the outcomes I want to look at. So have the the Fstatistics value and the pvalue for that....not sure if this is relevant. However, I want the effect of each individual intervention using cohen's D.
(If wanted to reference, the paper is pubmed ID# 28944527, and is also attached..I'm interested in analyzing some of the psychological questionnaire data...outcomes for body image measures)
Do you know which formula I would use? I'm confused because I'm not sure if I need to you the pooled SD or the individual SD of each group in this situation.
Thank you in advance!
The study design  participants are assigned to one of two groups (parallel design), but not a true RCT because both are interventional (no control group).
What I want to calculate  I want to include both interventions in my metaanalysis, and calculate an effect size for each of them (cohen's D). I am thinking I will compare the pre and post intervention scores for each group, for my effect sizes.
Information I have  I have pre and post means and standard deviations for each of the groups for my outcome measure. I also have the sample size of each group, of course. The paper, also reports the effect sizes (using ηp^2) for a 2 X 2 ANOVA (intervention groups  intervention1, intervention 2  BY time  baseline, post)...for the "time factor" of one of the outcomes I want to look at. So have the the Fstatistics value and the pvalue for that....not sure if this is relevant. However, I want the effect of each individual intervention using cohen's D.
(If wanted to reference, the paper is pubmed ID# 28944527, and is also attached..I'm interested in analyzing some of the psychological questionnaire data...outcomes for body image measures)
Do you know which formula I would use? I'm confused because I'm not sure if I need to you the pooled SD or the individual SD of each group in this situation.
Thank you in advance!
Attachments

441.4 KB Views: 6
Last edited: