[R-meta] vcalc type vs obs

[Viechtbauer, Wolfgang (NP)]

Hi Filippo,

It doesn't matter if you use 'obs' or 'type' in this case - the result will be the same (as you noted). The two arguments become relevant in case one wants to handle the case where multiple types of constructs have been assessed and in at least some studies the same construct was measured using multiple scales or instruments. Then one can specify 'type' and 'obs' separately and also specify different correlations.

Best,
Wolfgang

>-----Original Message-----
>From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-project.org] On
>Behalf Of Filippo Gambarota
>Sent: Monday, 22 August, 2022 9:59
>To: R meta
>Subject: [R-meta] vcalc type vs obs
>
>Hello!
>I'm running a multivariate model for the first time and I'm
>constructing the covariance matrix with vcalc. My situation is that I
>have n papers and each papers could have k effect size belonging to
>different measures but measured on the same pool of subjects.
>For this situation I'm using the vcalc function as follows:
>
>vcalc(vi, cluster = id, obs = outcome2, data = dat, rho = 0.7)
>
>Where cluster is the paper index and outcome2 is the outcome (as
>string) within each paper that of course, the same outcome could be
>present also in another paper.
>
>Is that correct? because carefully reading the documentation, maybe
>the argument type is more appropriate but the resulting matrix is the
>same.
>Thank you!
>
>--
>Filippo Gambarota
>PhD Student - University of Padova
>Department of Developmental and Social Psychology
>Website: filippogambarota
>Research Group: Colab   Psicostat