Re: [R] questions about metafor package
Hi, Emilie. For your second question. You may check Gleser and Olkin (2009). They gave several formulas to estimate the sampling covariance for dependent effect sizes. One of them can be applied in your case. Gleser, L. J., Olkin, I. (2009). Stochastically dependent effect sizes. In H. Cooper, L. V. Hedges, J. C. Valentine (Eds.), The handbook of research synthesis and meta-analysis. (2nd ed., pp. 357-376). New York: Russell Sage Foundation. Regards, Mike -- - Mike W.L. Cheung Phone: (65) 6516-3702 Department of Psychology Fax: (65) 6773-1843 National University of Singapore http://courses.nus.edu.sg/course/psycwlm/internet/ - On Sat, Aug 20, 2011 at 10:19 PM, Michael Dewey i...@aghmed.fsnet.co.ukwrote: At 16:21 17/08/2011, Emilie MAILLARD wrote: Hello, à I would like to do a meta-analysis with the package ëà metaforà û. Ideally I would like to use a mixed model because Iââ¬â¢m interested to see the effect of some moderators. But the data set I managed to collect from literature presents two limits. à -à à à à à à à à Firstly, for each observation, I have means for a treatment and for a control, but I donââ¬â¢t always have corresponding standard deviations (52 of a total of 93 observations donââ¬â¢t have standard deviations). Nevertheless I have the sample sizes for all observations so I wonder if it was possible to weight observations by sample size in the package ëà metaforà û. -à à à à à à à à Secondly, some observations are probably not independent as I have sometimes several relevant observations for a same design. More precisely, for these cases, the control mean is identical but treatment means varied. Ideally, I would not like to do a weighted average for these non-independent observations because these observations represent levels of a moderator. I know that the package ëà metaforà û is not designed for the analysis of correlated outcomes. What are the dangers of using the package even if observations are not really independent ? à Emilie, I am not sure whether this is the answer to your problem of observations which are not independent but you might also look at the metaSEM package http://courses.nus.edu.sg/**course/psycwlm/internet/**metaSEM/http://courses.nus.edu.sg/course/psycwlm/internet/metaSEM/ I am still trying to understand his paper on this (see link for reference) but he is trying to embed meta-analysis within the structural equation framework and it may be possible to cope with lack of independence in that way. But as I say I am still trying to come to grips with the paper. à Thank you for your help, à Ãâ°milie. [[alternative HTML version deleted]] Michael Dewey i...@aghmed.fsnet.co.uk http://www.aghmed.fsnet.co.uk/**home.htmlhttp://www.aghmed.fsnet.co.uk/home.html __** R-help@r-project.org mailing list https://stat.ethz.ch/mailman/**listinfo/r-helphttps://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/** posting-guide.html http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] questions about metafor package
At 16:21 17/08/2011, Emilie MAILLARD wrote: Hello,  I would like to do a meta-analysis with the package « metafor ». Ideally I would like to use a mixed model because Iâm interested to see the effect of some moderators. But the data set I managed to collect from literature presents two limits.  -        Firstly, for each observation, I have means for a treatment and for a control, but I donât always have corresponding standard deviations (52 of a total of 93 observations donât have standard deviations). Nevertheless I have the sample sizes for all observations so I wonder if it was possible to weight observations by sample size in the package « metafor ». -        Secondly, some observations are probably not independent as I have sometimes several relevant observations for a same design. More precisely, for these cases, the control mean is identical but treatment means varied. Ideally, I would not like to do a weighted average for these non-independent observations because these observations represent levels of a moderator. I know that the package « metafor » is not designed for the analysis of correlated outcomes. What are the dangers of using the package even if observations are not really independent ?  Emilie, I am not sure whether this is the answer to your problem of observations which are not independent but you might also look at the metaSEM package http://courses.nus.edu.sg/course/psycwlm/internet/metaSEM/ I am still trying to understand his paper on this (see link for reference) but he is trying to embed meta-analysis within the structural equation framework and it may be possible to cope with lack of independence in that way. But as I say I am still trying to come to grips with the paper.  Thank you for your help,  Ãmilie. [[alternative HTML version deleted]] Michael Dewey i...@aghmed.fsnet.co.uk http://www.aghmed.fsnet.co.uk/home.html __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
[R] questions about metafor package
Hello,  I would like to do a meta-analysis with the package « metafor ». Ideally I would like to use a mixed model because Iâm interested to see the effect of some moderators. But the data set I managed to collect from literature presents two limits.  -        Firstly, for each observation, I have means for a treatment and for a control, but I donât always have corresponding standard deviations (52 of a total of 93 observations donât have standard deviations). Nevertheless I have the sample sizes for all observations so I wonder if it was possible to weight observations by sample size in the package « metafor ». -        Secondly, some observations are probably not independent as I have sometimes several relevant observations for a same design. More precisely, for these cases, the control mean is identical but treatment means varied. Ideally, I would not like to do a weighted average for these non-independent observations because these observations represent levels of a moderator. I know that the package « metafor » is not designed for the analysis of correlated outcomes. What are the dangers of using the package even if observations are not really independent ?   Thank you for your help,  Ãmilie. [[alternative HTML version deleted]] __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] questions about metafor package
Dear Emilie, Regarding your questions: 1) It's not the weighting that is the main issue when you do not have the SDs. The problem is that you need the SDs to calculate the sampling variances of the mean differences (I assume that this is your outcome measure for the meta-analysis). Those are needed to calculate the standard errors of the model coefficients. There are two possible routes to take. The first would be to try your hardest to get your hands on as many of the missing SDs as possible. Whatever is left missing could be imputed, using a sensible range of values and checking for the robustness of the findings. The other approach would be to choose some other weights (e.g., sample size weights), then fit the model by WLS, and then estimate the standard errors of the model coefficients using a robust method (e.g., using a sandwich estimator). 2) Difficult to say. I haven’t had a chance to read this article, but this will probably tell you more: Ishak, K. J., Platt, R. W., Joseph, L., Hanley, J. A. (2008). Impact of approximating or ignoring within-study covariances in multivariate meta-analyses. Statistics in Medicine, 27(5), 670-686. Best, -- Wolfgang Viechtbauer Department of Psychiatry and Neuropsychology School for Mental Health and Neuroscience Maastricht University, P.O. Box 616 6200 MD Maastricht, The Netherlands Tel: +31 (43) 368-5248 Fax: +31 (43) 368-8689 Web: http://www.wvbauer.com -Original Message- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Emilie MAILLARD Sent: Wednesday, August 17, 2011 17:21 To: r-help@r-project.org Subject: [R] questions about metafor package Hello, I would like to do a meta-analysis with the package « metafor ». Ideally I would like to use a mixed model because I’m interested to see the effect of some moderators. But the data set I managed to collect from literature presents two limits. - Firstly, for each observation, I have means for a treatment and for a control, but I don’t always have corresponding standard deviations (52 of a total of 93 observations don’t have standard deviations). Nevertheless I have the sample sizes for all observations so I wonder if it was possible to weight observations by sample size in the package « metafor ». - Secondly, some observations are probably not independent as I have sometimes several relevant observations for a same design. More precisely, for these cases, the control mean is identical but treatment means varied. Ideally, I would not like to do a weighted average for these non-independent observations because these observations represent levels of a moderator. I know that the package « metafor » is not designed for the analysis of correlated outcomes. What are the dangers of using the package even if observations are not really independent ? Thank you for your help, Émilie. __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Re: [R] questions about metafor package
. - Firstly, for each observation, I have means for a treatment and for a control, but I don’t always have corresponding standard deviations (52 of a total of 93 observations don’t have standard deviations). Nevertheless I have the sample sizes for all observations so I wonder if it was possible to weight observations by sample size in the package « metafor ». Following what Wolfgang said, do you have some other information, such as p-values, or standard errors of the difference, or confidence intervals, which would allow you to calculate (or approximate) the pooled SD? jeremy __ R-help@r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
