[R-meta] Help with meta-analisys of variation

[Viechtbauer, Wolfgang (SP)]

The covariance between two lnSD values depends on the correlation between the two variables (for which you computed the lnSD values) and can be estimated with:

r^2 / (2*(n-1))

where r is the correlation between the two variables. So, you need to know the correlation between the measurements at different stages in order to compute the covariances.

Best,
Wolfgang

-----Original Message-----
From: Fernando Klitzke Borszcz [mailto:fernandoborszcz using gmail.com] 
Sent: Thursday, 18 July, 2019 14:35
To: Viechtbauer, Wolfgang (SP)
Cc: r-sig-meta-analysis using r-project.org
Subject: Re: [R-meta] Help with meta-analisys of variation

Dear Professor Wolfgang, 

Thank you for your response.

My question is if I should use any value of correlation to determine the variance-covariance matrix in this type of data? I am not sure if I could use a correlation value, because the variances of log-transformed standard deviations (lnSD) are based only on sample size. 

Below an example of my data.

    study    yi     vi exercise stage
1      A 2.351 0.0250      row     3
2      A 2.351 0.0250      row     4
3      A 2.261 0.0250      row     5
4      A 1.922 0.0250      row     7
5      A 1.887 0.0250      row    10
6      B 1.884 0.0313    cycle     3
7      B 1.851 0.0313    cycle     4
8      B 1.774 0.0313    cycle     7
9      B 1.740 0.0313    cycle    10
10     C 1.831 0.0278    cycle     3
11     D 1.732 0.0294      run     5

Em qui, 18 de jul de 2019 às 07:26, Viechtbauer, Wolfgang (SP) <wolfgang.viechtbauer using maastrichtuniversity.nl> escreveu:
Dear Fernando,

I am not sure I fully understand your question. My interpretation of your question/problem is as follows. You have studies that measure some property of interest in a group of a subjects at multiple timepoints. And for each timepoint, you compute ln(SD) with sampling variance 1/(2*(n-1)). But since multiple ln(SD) values are based on the same sample of subjects, you want to compute the covariance between those ln(SD) values and you are not sure how to do that. Is this the essence of your question?

Best,
Wolfgang

-----Original Message-----
From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-project.org] On Behalf Of Fernando Klitzke Borszcz
Sent: Thursday, 11 July, 2019 15:42
To: r-sig-meta-analysis using r-project.org
Subject: [R-meta] Help with meta-analisys of variation

Hi all,

I am conducting a meta-analysis about standard deviations of change between
2 conditions in the metafor package. I am using effect sizes (lnSD) and
their sampling variance, based on Nakagawa et al. (
https://besjournals.onlinelibrary.wiley.com/doi/full/10.1111/2041-210X.12309).
However, many studies included in my analysis, the same subject produces
multiple results (multiple time points). Thus, I should determine the
variance-covariance matrix of the effects. However, I am not sure if it is
appropriate to determine the variance-covariance matrix using any
correlation value for this type of effect, because effects variances are
based only on the sample size (1/2[n-1]). Is there any alternative to
determine the variance-covariance matrix for this type of data, or I should
just include the study as a random effect?

Best wishes!

Att.
Fernando Borszcz

-- 
Att.
Fernando Klitzke Borszcz
Doutorando em Educação Física
Laboratório de Esforço Físico - LAEF
Centro de Desportos - CDS
Universidade Federal de Santa Catarina- UFSC
(48) 998261406
http://lattes.cnpq.br/4442928023267586
orcid.org/0000-0002-3773-6906