Meta Regression
Because of the heterogeneity of the effects in my meta-analysis I wanted to look for moderators, using meta Regression.
I put my moderators in as factors for example:
-Randomisation (randomised vs. Non randomised)
-Age group (baby Vs child Vs teenager)
I'm only getting 1 estimate for the moderator with 2 steps (randomised is missing) and only 2 estimates for the moderators with 3 steps (baby is missing).
Why is that?
Thank you!
Comments
Hi kassiopeiia,
The moderators that you describe are categorical. A categorical variable is encoded with dummy variables. There are always one less dummy variables than the number of categories. Please consult a good book on regression analysis. I suggest Pituch & Stevens (2016) Applied Multivariate Statistics for the Social Sciences (https://www.routledge.com/Applied-Multivariate-Statistics-for-the-Social-Sciences-Analyses-with/Pituch-Stevens/p/book/9780415836661; but really any book that discusses the general linear model would also discuss this.)
Best of luck.