I'm conducting a review with a set of studies reporting only a primary outcome and many predictor variables. Is it possible to easily perform a metaregression to assess the effect of predictors on a single primary outcome, maybe with a free software?
Metafor in R is very flexible and easy to use. Avoiding R because it is not 'easy' to use is a bit silly in my view. It takes some getting used to but the wealth of packages available with the very latests statistical possibilities makes it more or less a prerequisite at least in many fields of Biology. Most packages include very detailed tutorials. Good luck.
Dear Stefano,
I would suggest you try the Comprehensive Meta-analysis. It has many analyses to perform, although you cannot 'twinker' with the options. And although it is not freeware, you can download it and run it for 10 uses (at each computer that you install it). Tip: When you open the program once, do not close it, unless you absolutely have to, as otherwise you will run out of your 10 uses quickly..
Also, there is Stata, which is less user-friendly, but capable to run many more analyses and individualize them to your needs, but you have to purchase a license.
Then there is also R, which is freeware, very strong and with many many possibilities, but it is difficult to use, so I would suggest avoiding it.
Hope it helped!
Hey Stefano
I searched far and wide on the internet for free meta-analytic software. I found that the free stuff was limited, not reliable, and outdated. I ended up purchasing Comprehensive Meta-Analysis, and I strongly recommend it. It's very user friendly and easy to use. Other programs such as Stata, R, and SPSS may be able to run a meta-regression, but you'd have to get the license first, then deal with complex syntax. If I were you, I'd purchase Comprehensive Metal-Analysis. As a researcher with interest in meta-analysis, I see it as an investment.
Best of luck!
You can try Wilson & Lipsey's SPSS macro available here: http://mason.gmu.edu/~dwilsonb/ma.html Its useful for both categorical & continuous moderators. I used it in combination with CMA 2.0 for a recent meta-analysis. But I've purchased CMA 3.0...I recommend it.
Dear Stefano,
The Meta-analyst software is a free software for meta-analysis with meta-regression option.
Meta-Analyst: software for meta-analysis of binary, continuous and diagnostic data. Wallace BC, Schmid CH, Lau J, Trikalinos TA. BMC Med Res Methodol. 2009 Dec 4;9:80. doi: 10.1186/1471-2288-9-80.
You can download the software freely.
Best wishes,
Ramin
Another free software for meta-analysis is Meta-Disc software which is specifically designed for diagnostic studies. Actually even Comprehensive Meta-analysis can not replace it due to peculiar nature of diagnostic meta-analyses.
Meta-DiSc: a software for meta-analysis of test accuracy data.
Zamora J, Abraira V, Muriel A, Khan K, Coomarasamy A.
BMC Med Res Methodol. 2006 Jul 12;6:31.
Ramin
Metafor in R is very flexible and easy to use. Avoiding R because it is not 'easy' to use is a bit silly in my view. It takes some getting used to but the wealth of packages available with the very latests statistical possibilities makes it more or less a prerequisite at least in many fields of Biology. Most packages include very detailed tutorials. Good luck.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Meta-Analysis