Hey there!
I have data that includes the following arms:
- Healthy individuals who did not receive the drug
- Ill individuals who did not receive the drug
- Ill individuals who did receive the drug
- Healthy individuals who did receive the drug
I want to perform a meta-analysis and show all my results for one outcome in a single forest plot. My effect size is SMD.
Can you tell me if it's possible and how to do it?
And, by the way, does Stata have my back? If yes, what command should I use?
Hey Chuck, thanks for sharing the paper but it doesn't really answer my question. It's mostly focused on the calculations and doesn't do a great job visually reporting the results. Like I mentioned in my original question, I'm trying to find a way to show the effect sizes in a more visual way.
A meta-analysis combines effect size from different studies and produces an overall effect size measure. You seem to have only one study and you want to compare groups.
The second feature of interest is that you have a two-way classification – that is, your groups are not separate categories. You have the four combinations of two binary variables : ill/well and drug/no drug. The questions would be
So there are in fact two predictor measures and their interaction.
There are various ways of displaying interactions like this (check out https://stats.oarc.ucla.edu/stata/seminars/interactions-stata/) that would be more helpful.
Stata does indeed have your back, but it can't tap you on the shoulder when you're facing in the wrong direction! ;—})
- Badges
- Science topic
- Similar topics
- Correction