Dear All,
According to the spm manual I have read: "Contrast of parameter estimates: SPM will prompt for a specific contrast (e.g., active>rest). The plot will show effect size and error bars (in %)." This gives you a plot with contrast on the x-axis and contrast estimate at [x,y,z] on the y-axis. I understand this to mean that your effect sizes are on the y-axis. My question is: how are these effect sizes generated and what kind are they? Are they the kind of effect sizes you can judge using a Cohen's D scale (ex. 1 being a big effect and 0.2 being a smaller effect?). I guess my general question is: How do you know if the effect sizes you are getting out of the plot are big or small?
Thank you,
Cyrus
|