[R-meta] Meta-analysis on positive and negative outcomes: Is it correct to take the reversed r values?

[Viechtbauer, Wolfgang (SP)]

Dear Gladys,

Yes, you can flip the sign. In essence, this is what would happen if a study where Y reflects depression had reverse-scored their depression measure (such that higher values reflect *less* depression).

Whether the association between X and 'positive' measures of mental health is the same as the association between X and 'negative' measures of mental health (after flipping the sign of the latter) is a different issue. With enough studies, one could code a moderator variable to indicate whether Y was originally a positive or negative measure and examine whether the association between X and Y differs across these two groups.

Best,
Wolfgang

>-----Original Message-----
>From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-project.org]
>On Behalf Of Gladys Barragan-Jason
>Sent: Monday, 18 January, 2021 15:33
>To: r-sig-meta-analysis using r-project.org
>Subject: [R-meta] Meta-analysis on positive and negative outcomes: Is it
>correct to take the reversed r values?
>
>Dear all,
>I have a question about taking the reversed values when combining different
>proxies of one factor. For instance, if you're doing  a meta-analysis on the
>impact of X on Y where Y is  mental health and you have several measures of
>Y such as depression (r= -0.4) , stress (r=-0.1) and emotional balance
>(r=+0.5-) and positive moods (r=+0.2), is it correct to transform (take to
>reversed values) for "negative health" (r=+0.4 and r=0.1) and pooled
>everything together or is it better to make to different meta-analysis (one
>on positive and one on negative effects), or maybe both?
>Thanks a lot for your response.
>Best,
>Gladys
>
>--
>------------------------------------------
>Gladys Barragan-Jason, PhD.  Website
>Station d'Ecologie Théorique et Expérimentale (SETE)
>CNRS de Moulis