[R-meta] Meta analysis with least squares mean (Change %) and SE
Hi every body. I want to do a meta-analysis of continuous data, comparing the percentage of change in the level of lipids, of a treatment vs placebo. The % of change is made based on least squares mean baseline and follow-up and is reported as %(SE) 1 - How can I perform the meta analysis of this data? MD? SMD? 2- I don't know the co-variables used for the LSM. Is this a problem ? 3 - in some studies% (SD) is reported, can they be analyzed together with% (SE)? 4- The %(SE) can be converted to %(DS) ? Thank you ML Lorenzo Mart?n Lobo MTSAC, FACC, FESC Especialista Jerarquizado en Cardiolog?a Jefe de Dpto Enf. Cardiovasculares y Cardiometabolismo Hospital Militar Campo de Mayo. Jefe de Cardiolog?a Hospital Militar Campo de Mayo Ex Jefe de Unidad Coronaria Hospital Militar Campo de Mayo Miembro Titular de la Sociedad Argentina de Cardiolog?a Fellow American College of Cardiology Fellow European Society of Cardiology Ex Miembro del Area de Investigaci?n de la SAC Ex Director del Consejo de Aterosclerosis y Trombosis de la SAC Miembro Asesor del Consejo de Aterosclerosis y Trombosis de la SAC Ex Director del Consejo de Epidemiolog?a y Prevenci?n Cardiovascular de la SAC Miembro Asesor del Consejo de Epidemiolog?a y Prevenci?n Cardiovascular de la SAC Experto en Lipidos de la Sociedad Argentina de Lipidos. Miembro de la Sociedad Argentina de Lipidos. Instructor de ACLS de la American Heart Association
De: Dr. Gerta R?cker <ruecker at imbi.uni-freiburg.de>
Enviado: domingo, 26 de abril de 2020 15:13 Para: Martin Lobo <mlobo4370 at hotmail.com>; r-sig-meta-analysis at r-project.org <r-sig-meta-analysis at r-project.org> Asunto: Re: [R-meta] meta analysis with standard deviation or standard errors Hi Martin given you have the two studies you gave as an example, and given their sample sizes (you did not provide, thus I take both to have n = 50 here), with R package meta, function metamean() this goes like: library(meta) m1 <- metamean(n = c(50,50), mean = c(-23,-20), sd = c(12,9)) m1 # Print results forest(m1) # Forest plot Best, Gerta Am 26.04.2020 um 06:30 schrieb Martin Lobo: Hi Gerta for example: I have the difference between ldl after a treatment of 5 studies ( and its standard deviation ). The studies are paired data. study 1 difference -23(+-12), study 2-20(+-9), etc. I can use ml1=0 and ml2=difference, sl1=0 and sl2=sd difference, and use MD for indepent data ? Lorenzo Mart?n Lobo MTSAC, FACC, FESC Especialista Jerarquizado en Cardiolog?a Jefe de Dpto Enf. Cardiovasculares y Cardiometabolismo Hospital Militar Campo de Mayo. Jefe de Cardiolog?a Hospital Militar Campo de Mayo Ex Jefe de Unidad Coronaria Hospital Militar Campo de Mayo Miembro Titular de la Sociedad Argentina de Cardiolog?a Fellow American College of Cardiology Fellow European Society of Cardiology Ex Miembro del Area de Investigaci?n de la SAC Ex Director del Consejo de Aterosclerosis y Trombosis de la SAC Miembro Asesor del Consejo de Aterosclerosis y Trombosis de la SAC Ex Director del Consejo de Epidemiolog?a y Prevenci?n Cardiovascular de la SAC Miembro Asesor del Consejo de Epidemiolog?a y Prevenci?n Cardiovascular de la SAC Experto en Lipidos de la Sociedad Argentina de Lipidos. Miembro de la Sociedad Argentina de Lipidos. Instructor de ACLS de la American Heart Association ________________________________ De: Gerta Ruecker <ruecker at imbi.uni-freiburg.de><mailto:ruecker at imbi.uni-freiburg.de> Enviado: viernes, 24 de abril de 2020 05:13 Para: Martin Lobo <mlobo4370 at hotmail.com><mailto:mlobo4370 at hotmail.com>; r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> <r-sig-meta-analysis at r-project.org><mailto:r-sig-meta-analysis at r-project.org> Asunto: Re: [R-meta] meta analysis with standard deviation or standard errors Dear Martin, I'm not sure I understand this question. You want to do a meta-analysis of pre-post differences. If you have the mean pre-post differences with their standard deviations, this is what you want to pool, so what is the problem? What do you mean by "independent sample method" and what is 0? Best, Gerta Am 22.04.2020 um 20:44 schrieb Martin Lobo: Gerta: If the studies report the mean differences with the standard deviations (pre-post), I can use the independent sample method with these data, using 0 for a mean and its deviation. thanks Lorenzo Mart?n Lobo MTSAC, FACC, FESC Especialista Jerarquizado en Cardiolog?a Jefe de Dpto Enf. Cardiovasculares y Cardiometabolismo Hospital Militar Campo de Mayo. Jefe de Cardiolog?a Hospital Militar Campo de Mayo Ex Jefe de Unidad Coronaria Hospital Militar Campo de Mayo Miembro Titular de la Sociedad Argentina de Cardiolog?a Fellow American College of Cardiology Fellow European Society of Cardiology Ex Miembro del Area de Investigaci?n de la SAC Ex Director del Consejo de Aterosclerosis y Trombosis de la SAC Miembro Asesor del Consejo de Aterosclerosis y Trombosis de la SAC Ex Director del Consejo de Epidemiolog?a y Prevenci?n Cardiovascular de la SAC Miembro Asesor del Consejo de Epidemiolog?a y Prevenci?n Cardiovascular de la SAC Experto en Lipidos de la Sociedad Argentina de Lipidos. Miembro de la Sociedad Argentina de Lipidos. Instructor de ACLS de la American Heart Association ________________________________ De: Gerta Ruecker <ruecker at imbi.uni-freiburg.de><mailto:ruecker at imbi.uni-freiburg.de> Enviado: mi?rcoles, 22 de abril de 2020 11:16 Para: Martin Lobo <mlobo4370 at hotmail.com><mailto:mlobo4370 at hotmail.com>; r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> <r-sig-meta-analysis at r-project.org><mailto:r-sig-meta-analysis at r-project.org> Asunto: Re: [R-meta] meta analysis with standard deviation or standard errors For paired samples (changes, differences) I would preferably use MD. Gerta Am 22.04.2020 um 16:02 schrieb Martin Lobo: THANK YOU GERTA !!! Ok For individual samples I use MD or SMD, depending on whether the measurements are on the same scale or not. for paired samples, should i use MC or SMCC? Regard Lorenzo Mart?n Lobo MTSAC, FACC, FESC Especialista Jerarquizado en Cardiolog?a Jefe de Dpto Enf. Cardiovasculares y Cardiometabolismo Hospital Militar Campo de Mayo. Jefe de Cardiolog?a Hospital Militar Campo de Mayo Ex Jefe de Unidad Coronaria Hospital Militar Campo de Mayo Miembro Titular de la Sociedad Argentina de Cardiolog?a Fellow American College of Cardiology Fellow European Society of Cardiology Ex Miembro del Area de Investigaci?n de la SAC Ex Director del Consejo de Aterosclerosis y Trombosis de la SAC Miembro Asesor del Consejo de Aterosclerosis y Trombosis de la SAC Ex Director del Consejo de Epidemiolog?a y Prevenci?n Cardiovascular de la SAC Miembro Asesor del Consejo de Epidemiolog?a y Prevenci?n Cardiovascular de la SAC Experto en Lipidos de la Sociedad Argentina de Lipidos. Miembro de la Sociedad Argentina de Lipidos. Instructor de ACLS de la American Heart Association ________________________________ De: Gerta Ruecker <ruecker at imbi.uni-freiburg.de><mailto:ruecker at imbi.uni-freiburg.de> Enviado: mi?rcoles, 22 de abril de 2020 10:26 Para: Martin Lobo <mlobo4370 at hotmail.com><mailto:mlobo4370 at hotmail.com>; r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> <r-sig-meta-analysis at r-project.org><mailto:r-sig-meta-analysis at r-project.org> Asunto: Re: [R-meta] meta analysis with standard deviation or standard errors Dear Martin, I am not sure whether I understand you correctly, but I see the following cases: 1. You have pre-post changes (differences) with sd (or se) of these changes -> you can use these for pooling 2. You have pre values and post values and their intra-individual correlations (not frequently the case) -> you can use the correlations to calculate the sd/se for the differences (and then pool as in case 1) 3. You have pre values and post values, but no correlations and no sd or se for the differences -> you might impute a correlation and proceed as in case 2 4. You can also mix pre-post changes and post values, but only for mean differences, not for standardized mean differences, see Cochrane Handbook https://training.cochrane.org/handbook/current/chapter-10#section-10-5-2<https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Ftraining.cochrane.org%2Fhandbook%2Fcurrent%2Fchapter-10%23section-10-5-2&data=02%7C01%7C%7C4f3ff2aa366e45358eb308d7ea0d74b9%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637235216027108324&sdata=gp%2BzGz3pp6l1d%2FD8qWFy77ZRVATZ9YQsLnUvnj62EuU%3D&reserved=0> Best, Gerta Am 22.04.2020 um 14:30 schrieb Martin Lobo: Dear Gerta, thank tou very mucha for tour time. 1- the MC and SMCC are the methods I found for paired samples on page 103 of the metafor manual, I understood that they were equivalent to the MD and SMD of the individual samples. Manual Link: https://cran.r-project.org/web/packages/metafor/metafor.pdf<https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fcran.r-project.org%2Fweb%2Fpackages%2Fmetafor%2Fmetafor.pdf&data=02%7C01%7C%7C4f3ff2aa366e45358eb308d7ea0d74b9%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637235216027118317&sdata=pFX4j9XTOBT1c%2FBlD3n3g55gvxtDuzNOJMwohTN4uHk%3D&reserved=0> If I had the pre post standard averages and deviations, only the difference with your standard deviation, would I no longer need the ri? In that case what method do I use or what code? thank you so much Martin Lorenzo Mart?n Lobo MTSAC, FACC, FESC Especialista Jerarquizado en Cardiolog?a Jefe de Dpto Enf. Cardiovasculares y Cardiometabolismo Hospital Militar Campo de Mayo. Jefe de Cardiolog?a Hospital Militar Campo de Mayo Ex Jefe de Unidad Coronaria Hospital Militar Campo de Mayo Miembro Titular de la Sociedad Argentina de Cardiolog?a Fellow American College of Cardiology Fellow European Society of Cardiology Ex Miembro del Area de Investigaci?n de la SAC Ex Director del Consejo de Aterosclerosis y Trombosis de la SAC Miembro Asesor del Consejo de Aterosclerosis y Trombosis de la SAC Ex Director del Consejo de Epidemiolog?a y Prevenci?n Cardiovascular de la SAC Miembro Asesor del Consejo de Epidemiolog?a y Prevenci?n Cardiovascular de la SAC Experto en Lipidos de la Sociedad Argentina de Lipidos. Miembro de la Sociedad Argentina de Lipidos. Instructor de ACLS de la American Heart Association ________________________________ De: Gerta Ruecker <ruecker at imbi.uni-freiburg.de><mailto:ruecker at imbi.uni-freiburg.de> Enviado: martes, 21 de abril de 2020 14:44 Para: Martin Lobo <mlobo4370 at hotmail.com><mailto:mlobo4370 at hotmail.com>; r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> <r-sig-meta-analysis at r-project.org><mailto:r-sig-meta-analysis at r-project.org> Asunto: Re: [R-meta] meta analysis with standard deviation or standard errors Dear Martin, Sorry for the delay. The problem is that the mean and sd of pre and post do not suffice to know the sd of the pairwise differences, except one makes some assumptions about the intraindividual pre-post correlation. See the attached R code PrePost.R for illustration. Do you mean by ri the correlation coefficients? If you impute them (say, 0.5), you may analyse the pre-post changes, but you should have some (external) evidence for using a certain value. I am not sure about each one of your 5 points below, see inline below. Best, Gerta Am 17.04.2020 um 14:22 schrieb Martin Lobo: Thank you very much Gerta. I asked the question to see how I can solve two problems I have. 1- If I want to do an metaanalysis of mean difference analysis (Paired data, pre-post) I have mean and sd pre and post, what methodd i use, MC , SMCC, etc What is MC, SMCC? I don't know for what these abbreviations stand. Otherwise, see above. 2- If I only have the mean and standard deviation as I do See above. 3 - ri is always necessary or can be imputed in some way See also above 4 - without ri the standard deviation of the mean difference can be estimated Not without knowing or making assumptions about the correlation, as said above. 5 - regarding question 4, both for independent samples and for paired samples For independent samples it is different, because for differences of independent means we have: sd(X + Y) = sqrt(var(X + Y)) = sqrt(var(X) + var(Y)) = sqrt(sd(X)^2 + sd(Y)^2) For paired (more general. correlated) variables: sd(X + Y) = sqrt(var(X) + var(Y) - 2Cov(X,Y)) ________________________________ De: Gerta Ruecker <ruecker at imbi.uni-freiburg.de><mailto:ruecker at imbi.uni-freiburg.de> Enviado: viernes, 17 de abril de 2020 08:12 Para: Martin Lobo <mlobo4370 at hotmail.com><mailto:mlobo4370 at hotmail.com>; r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> <r-sig-meta-analysis at r-project.org><mailto:r-sig-meta-analysis at r-project.org> Asunto: Re: [R-meta] meta analysis with standard deviation or standard errors Dear Martin, The answer is no. The standard error is not a measure of dispersion of the data, but a measure of the imprecision of estimation. A large standard error may come from large variability between data, but also from small sample size. The standard error becomes always small if the sample size becomes large (law of large numbers). Best, Gerta Am 17.04.2020 um 13:07 schrieb Martin Lobo: Hello everyone ! I wanted to know if it is possible to use the standard error instead of the standard deviation as a measure of dispersion. using the MD or SMD method for independent samples. If this is possible, there would be some difference in the conclusions. Thank you so much Lorenzo Mart?n Lobo MTSAC, FACC, FESC Especialista Jerarquizado en Cardiolog?a Jefe de Dpto Enf. Cardiovasculares y Cardiometabolismo Hospital Militar Campo de Mayo. Jefe de Cardiolog?a Hospital Militar Campo de Mayo Ex Jefe de Unidad Coronaria Hospital Militar Campo de Mayo Miembro Titular de la Sociedad Argentina de Cardiolog?a Fellow American College of Cardiology Fellow European Society of Cardiology Ex Miembro del Area de Investigaci?n de la SAC Ex Director del Consejo de Aterosclerosis y Trombosis de la SAC Miembro Asesor del Consejo de Aterosclerosis y Trombosis de la SAC Ex Director del Consejo de Epidemiolog?a y Prevenci?n Cardiovascular de la SAC Miembro Asesor del Consejo de Epidemiolog?a y Prevenci?n Cardiovascular de la SAC Experto en Lipidos de la Sociedad Argentina de Lipidos. Miembro de la Sociedad Argentina de Lipidos. Instructor de ACLS de la American Heart Association ________________________________ De: R-sig-meta-analysis <r-sig-meta-analysis-bounces at r-project.org><mailto:r-sig-meta-analysis-bounces at r-project.org> en nombre de r-sig-meta-analysis-request at r-project.org<mailto:r-sig-meta-analysis-request at r-project.org> <r-sig-meta-analysis-request at r-project.org><mailto:r-sig-meta-analysis-request at r-project.org> Enviado: mi?rcoles, 15 de abril de 2020 07:00 Para: r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> <r-sig-meta-analysis at r-project.org><mailto:r-sig-meta-analysis at r-project.org> Asunto: R-sig-meta-analysis Digest, Vol 35, Issue 8 Send R-sig-meta-analysis mailing list submissions to r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> To subscribe or unsubscribe via the World Wide Web, visit https://nam01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-sig-meta-analysis&data=02%7C01%7C%7C7f93a72da7b64707fe6d08d7e12439ac%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637225418004815037&sdata=Wed2UnN%2FV4z79%2Bb555NuEz7%2Fs9ta97aXHc18%2BxjrLLk%3D&reserved=0<https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-sig-meta-analysis&data=02%7C01%7C%7C4f3ff2aa366e45358eb308d7ea0d74b9%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637235216027128313&sdata=l5K69R0Nsm77OZ40LZCMHk2qBmppBx1Qfk7T7%2FFEago%3D&reserved=0> or, via email, send a message with subject or body 'help' to r-sig-meta-analysis-request at r-project.org<mailto:r-sig-meta-analysis-request at r-project.org> You can reach the person managing the list at r-sig-meta-analysis-owner at r-project.org<mailto:r-sig-meta-analysis-owner at r-project.org> When replying, please edit your Subject line so it is more specific than "Re: Contents of R-sig-meta-analysis digest..." Today's Topics: 1. Re: Dear Wolfgang (Viechtbauer, Wolfgang (SP)) 2. Re: Dear Wolfgang (Ju Lee) ---------------------------------------------------------------------- Message: 1 Date: Tue, 14 Apr 2020 20:43:51 +0000 From: "Viechtbauer, Wolfgang (SP)" <wolfgang.viechtbauer at maastrichtuniversity.nl><mailto:wolfgang.viechtbauer at maastrichtuniversity.nl> To: Ju Lee <juhyung2 at stanford.edu><mailto:juhyung2 at stanford.edu>, "r-sig-meta-analysis at r-project.org"<mailto:r-sig-meta-analysis at r-project.org> <r-sig-meta-analysis at r-project.org><mailto:r-sig-meta-analysis at r-project.org> Subject: Re: [R-meta] Dear Wolfgang Message-ID: <b411740819d1411da87d505cdeceb3e6 at UM-MAIL3214.unimaas.nl><mailto:b411740819d1411da87d505cdeceb3e6 at UM-MAIL3214.unimaas.nl> Content-Type: text/plain; charset="iso-8859-1" Yes, if the effect size measure is the same, one can make such a comparison. Also, there should not be any overlap in the studies included in the two meta-analyses (as otherwise the two estimates are not independent, as assumed by the test). And yes, you don't need sample sizes or tau^2 values or anything else - just the two estimates and their corresponding standard errors. And it doesn't depend on what random effects structure was used in the two meta-analyses -- assuming that the structures used in the two meta-analyses were appropriate for the studies at hand. Best, Wolfgang -----Original Message----- From: Ju Lee [mailto:juhyung2 at stanford.edu] Sent: Tuesday, 14 April, 2020 18:54 To: Viechtbauer, Wolfgang (SP); r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> Subject: Re: Dear Wolfgang Dear Wolfgang, Thanks for your insights. I am reaching out to my colleagues to see how they have made such transformation. In the meantime, based on the information that you have sent, it is possible to compare two different meta-analyses if they are using the same effect size, say lnRR? and this wald-type test can be performed only with grand mean effect sizes and their standard error, without sample sizes or tau value, if I understood correctly? How would this approach be actually applicable to publications that seemingly used similar mixed-effect models but there is no guarantee that random effect structures are standardized between the two? [[elided Hotmail spam]] Best, JU ________________________________________ From: Viechtbauer, Wolfgang (SP) <wolfgang.viechtbauer at maastrichtuniversity.nl><mailto:wolfgang.viechtbauer at maastrichtuniversity.nl> Sent: Tuesday, April 14, 2020 7:04 AM To: Ju Lee <juhyung2 at stanford.edu><mailto:juhyung2 at stanford.edu>; r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> <r- sig-meta-analysis at r-project.org<mailto:sig-meta-analysis at r-project.org>> Subject: RE: Dear Wolfgang Dear Ju, In principle, this might be of interest to you: https://nam01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.metafor-project.org%2Fdoku.php%2Ftips%3Acomp_two_independent_estimates&data=02%7C01%7C%7C7f93a72da7b64707fe6d08d7e12439ac%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637225418004815037&sdata=Tqgh0WpvUo70JTaihNWcZcbVQCQRpbprCYAxGKtlBGY%3D&reserved=0<https://eur02.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.metafor-project.org%2Fdoku.php%2Ftips%3Acomp_two_independent_estimates&data=02%7C01%7C%7C4f3ff2aa366e45358eb308d7ea0d74b9%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637235216027138306&sdata=albXcrZ%2FnriCs0Qzh6RFiQcKmsi7KB78DKXCFbk62ec%3D&reserved=0> However, a standardized mean difference is given by (m1-m2)/sd, while a (log) response ratio is log(m1/m2). I see no sensible way of converting the former to the later. Best, Wolfgang -----Original Message----- From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces at r- project.org] On Behalf Of Ju Lee Sent: Monday, 13 April, 2020 22:47 To: r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> Subject: [R-meta] Dear Wolfgang Dear Wolfgang, I hope you are doing well. My research group is currently working on a project where they are trying to compare effect sizes generated from their current mixed-effect meta- analysis with effect sizes (based on similar response variables) calculated in other meta-analysis publications. We are currently using log response ratio and are trying to make some statement or analysis to compare our grand mean effect sizes with other studies. In more details, we are examining how herbivorous animal control plant growth in degraded environment. Now, there is already a meta-analysis out there that has examined this (in comparable manner) in natural environment as opposed to our study. My colleagues want to know if there is a way to make some type of comparison (ex. whether responses are stronger in degraded vs. natural environemnts) between two effect sizes from these different studies using statistical approaches. So far what they have from other meta-analysis publication is grand mean hedges'd and var which they transformed to lnRR and var in hopes to compare with our lnRR effect sizes. My view is that this is not possible unless we can have their actual raw dataset and run a whole new model combining with our original raw dataset. But I wanted to reach out to you and the community if there is an alternative approaches to compare mean effect sizes among different meta- analysis which are assumed to have used similar approaches in study selection and models (another issue being different random effect structures used in different meta-analysis which may not be very apparent from method description). [[elided Hotmail spam]] Best, JU ------------------------------ Message: 2 Date: Wed, 15 Apr 2020 05:33:16 +0000 From: Ju Lee <juhyung2 at stanford.edu><mailto:juhyung2 at stanford.edu> To: "Viechtbauer, Wolfgang (SP)" <wolfgang.viechtbauer at maastrichtuniversity.nl><mailto:wolfgang.viechtbauer at maastrichtuniversity.nl>, "r-sig-meta-analysis at r-project.org"<mailto:r-sig-meta-analysis at r-project.org> <r-sig-meta-analysis at r-project.org><mailto:r-sig-meta-analysis at r-project.org> Subject: Re: [R-meta] Dear Wolfgang Message-ID: <BYAPR02MB5559407370455A06F0B047A8F7DB0 at BYAPR02MB5559.namprd02.prod.outlook.com><mailto:BYAPR02MB5559407370455A06F0B047A8F7DB0 at BYAPR02MB5559.namprd02.prod.outlook.com> Content-Type: text/plain; charset="utf-8" Dear Wolfgang, [[elided Hotmail spam]] I am not sure how my colleagues have transformed hedges' d to lnRR, based on what sources, but I will reach out again once I have more details. I, too, have not known if there is a way to make such effect size transformation. Thank you very much! Best wishes, JU ________________________________ From: Viechtbauer, Wolfgang (SP) <wolfgang.viechtbauer at maastrichtuniversity.nl><mailto:wolfgang.viechtbauer at maastrichtuniversity.nl> Sent: Tuesday, April 14, 2020 1:43 PM To: Ju Lee <juhyung2 at stanford.edu><mailto:juhyung2 at stanford.edu>; r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> <r-sig-meta-analysis at r-project.org><mailto:r-sig-meta-analysis at r-project.org> Subject: RE: Dear Wolfgang Yes, if the effect size measure is the same, one can make such a comparison. Also, there should not be any overlap in the studies included in the two meta-analyses (as otherwise the two estimates are not independent, as assumed by the test). And yes, you don't need sample sizes or tau^2 values or anything else - just the two estimates and their corresponding standard errors. And it doesn't depend on what random effects structure was used in the two meta-analyses -- assuming that the structures used in the two meta-analyses were appropriate for the studies at hand. Best, Wolfgang -----Original Message----- From: Ju Lee [mailto:juhyung2 at stanford.edu] Sent: Tuesday, 14 April, 2020 18:54 To: Viechtbauer, Wolfgang (SP); r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> Subject: Re: Dear Wolfgang Dear Wolfgang, Thanks for your insights. I am reaching out to my colleagues to see how they have made such transformation. In the meantime, based on the information that you have sent, it is possible to compare two different meta-analyses if they are using the same effect size, say lnRR? and this wald-type test can be performed only with grand mean effect sizes and their standard error, without sample sizes or tau value, if I understood correctly? How would this approach be actually applicable to publications that seemingly used similar mixed-effect models but there is no guarantee that random effect structures are standardized between the two? [[elided Hotmail spam]] Best, JU ________________________________________ From: Viechtbauer, Wolfgang (SP) <wolfgang.viechtbauer at maastrichtuniversity.nl><mailto:wolfgang.viechtbauer at maastrichtuniversity.nl> Sent: Tuesday, April 14, 2020 7:04 AM To: Ju Lee <juhyung2 at stanford.edu><mailto:juhyung2 at stanford.edu>; r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> <r- sig-meta-analysis at r-project.org<mailto:sig-meta-analysis at r-project.org>> Subject: RE: Dear Wolfgang Dear Ju, In principle, this might be of interest to you: https://nam01.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.metafor-project.org%2Fdoku.php%2Ftips%3Acomp_two_independent_estimates&data=02%7C01%7C%7C7f93a72da7b64707fe6d08d7e12439ac%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637225418004815037&sdata=Tqgh0WpvUo70JTaihNWcZcbVQCQRpbprCYAxGKtlBGY%3D&reserved=0<https://eur02.safelinks.protection.outlook.com/?url=http%3A%2F%2Fwww.metafor-project.org%2Fdoku.php%2Ftips%3Acomp_two_independent_estimates&data=02%7C01%7C%7C4f3ff2aa366e45358eb308d7ea0d74b9%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637235216027148305&sdata=oM1apNSzcsREkBZrDENqM%2Bbsp18Sq1Vcmb4fDvbFWo4%3D&reserved=0> However, a standardized mean difference is given by (m1-m2)/sd, while a (log) response ratio is log(m1/m2). I see no sensible way of converting the former to the later. Best, Wolfgang -----Original Message----- From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces at r- project.org] On Behalf Of Ju Lee Sent: Monday, 13 April, 2020 22:47 To: r-sig-meta-analysis at r-project.org<mailto:r-sig-meta-analysis at r-project.org> Subject: [R-meta] Dear Wolfgang Dear Wolfgang, I hope you are doing well. My research group is currently working on a project where they are trying to compare effect sizes generated from their current mixed-effect meta- analysis with effect sizes (based on similar response variables) calculated in other meta-analysis publications. We are currently using log response ratio and are trying to make some statement or analysis to compare our grand mean effect sizes with other studies. In more details, we are examining how herbivorous animal control plant growth in degraded environment. Now, there is already a meta-analysis out there that has examined this (in comparable manner) in natural environment as opposed to our study. My colleagues want to know if there is a way to make some type of comparison (ex. whether responses are stronger in degraded vs. natural environemnts) between two effect sizes from these different studies using statistical approaches. So far what they have from other meta-analysis publication is grand mean hedges'd and var which they transformed to lnRR and var in hopes to compare with our lnRR effect sizes. My view is that this is not possible unless we can have their actual raw dataset and run a whole new model combining with our original raw dataset. But I wanted to reach out to you and the community if there is an alternative approaches to compare mean effect sizes among different meta- analysis which are assumed to have used similar approaches in study selection and models (another issue being different random effect structures used in different meta-analysis which may not be very apparent from method description). [[elided Hotmail spam]] Best, JU [[alternative HTML version deleted]] ------------------------------ Subject: Digest Footer _______________________________________________ R-sig-meta-analysis mailing list R-sig-meta-analysis at r-project.org<mailto:R-sig-meta-analysis at r-project.org> https://nam01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-sig-meta-analysis&data=02%7C01%7C%7C7f93a72da7b64707fe6d08d7e12439ac%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637225418004815037&sdata=Wed2UnN%2FV4z79%2Bb555NuEz7%2Fs9ta97aXHc18%2BxjrLLk%3D&reserved=0<https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-sig-meta-analysis&data=02%7C01%7C%7C4f3ff2aa366e45358eb308d7ea0d74b9%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637235216027158300&sdata=M83Fp1ZOAoHzsAsx7iaoIRcKfPbRXjtRiUM1OgafsSg%3D&reserved=0> ------------------------------ End of R-sig-meta-analysis Digest, Vol 35, Issue 8 ************************************************** [[alternative HTML version deleted]] _______________________________________________ R-sig-meta-analysis mailing list R-sig-meta-analysis at r-project.org<mailto:R-sig-meta-analysis at r-project.org> https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis<https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fstat.ethz.ch%2Fmailman%2Flistinfo%2Fr-sig-meta-analysis&data=02%7C01%7C%7C4f3ff2aa366e45358eb308d7ea0d74b9%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637235216027168294&sdata=3bdxLa5tV%2Bt3i0ZMDmfztHUZyvpyKtPHEGEMBur6OMw%3D&reserved=0> -- Dr. rer. nat. Gerta R?cker, Dipl.-Math. Institute of Medical Biometry and Statistics, Faculty of Medicine and Medical Center - University of Freiburg Stefan-Meier-Str. 26, D-79104 Freiburg, Germany Phone: +49/761/203-6673 Fax: +49/761/203-6680 Mail: ruecker at imbi.uni-freiburg.de<mailto:ruecker at imbi.uni-freiburg.de> Homepage: https://www.imbi.uni-freiburg.de/persons/ruecker/person_view<https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.imbi.uni-freiburg.de%2Fpersons%2Fruecker%2Fperson_view&data=02%7C01%7C%7C4f3ff2aa366e45358eb308d7ea0d74b9%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637235216027178288&sdata=je5DL0ZOgCCtHsroH%2BG6mSzZPjNCvgQX8NQ1qu3FJ8w%3D&reserved=0> -- Dr. rer. nat. Gerta R?cker, Dipl.-Math. Institute of Medical Biometry and Statistics, Faculty of Medicine and Medical Center - University of Freiburg Stefan-Meier-Str. 26, D-79104 Freiburg, Germany Phone: +49/761/203-6673 Fax: +49/761/203-6680 Mail: ruecker at imbi.uni-freiburg.de<mailto:ruecker at imbi.uni-freiburg.de> Homepage: https://www.imbi.uni-freiburg.de/persons/ruecker/person_view<https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.imbi.uni-freiburg.de%2Fpersons%2Fruecker%2Fperson_view&data=02%7C01%7C%7C4f3ff2aa366e45358eb308d7ea0d74b9%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637235216027188282&sdata=QbVT75HpvYlBS5qD77w1j7xW9QRvKYMtv2g0vdbp4rE%3D&reserved=0> -- Dr. rer. nat. Gerta R?cker, Dipl.-Math. Institute of Medical Biometry and Statistics, Faculty of Medicine and Medical Center - University of Freiburg Stefan-Meier-Str. 26, D-79104 Freiburg, Germany Phone: +49/761/203-6673 Fax: +49/761/203-6680 Mail: ruecker at imbi.uni-freiburg.de<mailto:ruecker at imbi.uni-freiburg.de> Homepage: https://www.imbi.uni-freiburg.de/persons/ruecker/person_view<https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.imbi.uni-freiburg.de%2Fpersons%2Fruecker%2Fperson_view&data=02%7C01%7C%7C4f3ff2aa366e45358eb308d7ea0d74b9%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637235216027198282&sdata=dX4ierrCq65BD3RDuZ8VsgmET%2BLPy17o6Mrp9at916w%3D&reserved=0> -- Dr. rer. nat. Gerta R?cker, Dipl.-Math. Institute of Medical Biometry and Statistics, Faculty of Medicine and Medical Center - University of Freiburg Stefan-Meier-Str. 26, D-79104 Freiburg, Germany Phone: +49/761/203-6673 Fax: +49/761/203-6680 Mail: ruecker at imbi.uni-freiburg.de<mailto:ruecker at imbi.uni-freiburg.de> Homepage: https://www.imbi.uni-freiburg.de/persons/ruecker/person_view<https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.imbi.uni-freiburg.de%2Fpersons%2Fruecker%2Fperson_view&data=02%7C01%7C%7C4f3ff2aa366e45358eb308d7ea0d74b9%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637235216027208274&sdata=aBjmMjLQnIl2ekyfXZBsfD86wcq%2B%2Fys6set%2B1EaPIwo%3D&reserved=0> -- Dr. rer. nat. Gerta R?cker, Dipl.-Math. Institute of Medical Biometry and Statistics, Faculty of Medicine and Medical Center - University of Freiburg Stefan-Meier-Str. 26, D-79104 Freiburg, Germany Phone: +49/761/203-6673 Fax: +49/761/203-6680 Mail: ruecker at imbi.uni-freiburg.de<mailto:ruecker at imbi.uni-freiburg.de> Homepage: https://www.imbi.uni-freiburg.de/persons/ruecker/person_view<https://eur02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.imbi.uni-freiburg.de%2Fpersons%2Fruecker%2Fperson_view&data=02%7C01%7C%7C4f3ff2aa366e45358eb308d7ea0d74b9%7C84df9e7fe9f640afb435aaaaaaaaaaaa%7C1%7C0%7C637235216027218269&sdata=%2BA0Z1sGQjknVIX4jHGjVgIae9xvy7YlEagkGf07cPMg%3D&reserved=0>