[R-meta] predictors of longitudinal outcomes

Dear Catia,

Wouldn't it be more straightforward to address such research questions
using meta-analytic structural equation modeling? One could connect
antecedents (e.g., gender, SES) with skills at T2 or even T3 if there
are some studies with multiple measurement occasions. One could also
compare different models to examine the relevance of specific predictors.


Some resources in case you're not familiar with this approach:

Shiny app: https://sjak.shinyapps.io/webMASEM/

Video demonstration: https://www.youtube.com/watch?v=0v-CdNLa_eo

Article: Jak, S., Li, H., Kolbe, L., de Jonge, H., & Cheung, M. W. L.
(2021). Meta?analytic structural equation modeling made easy: A tutorial
and web application for one?stage MASEM. Research synthesis methods,
12(5), 590-606. https://doi.org/10.1002/jrsm.1498

The shiny app is based on the metaSEM package, which enables further
analyses within R:
https://cran.r-project.org/web/packages/metaSEM/vignettes/Examples.html



Best,
Lukasz
--
Lukasz Stasielowicz
Osnabr?ck University
Institute for Psychology
Research methods, psychological assessment, and evaluation
Seminarstra?e 20
49074 Osnabr?ck (Germany)

On 19.10.2022 12:00, r-sig-meta-analysis-request at r-project.org wrote:

Send R-sig-meta-analysis mailing list submissions to
      r-sig-meta-analysis at r-project.org

To subscribe or unsubscribe via the World Wide Web, visit

CACw+Tff0YnRmsFK8YXqe_dqn1EB16KSD6jt+v_e-dHf4sL6_-g at mail.gmail.com>

Content-Type: text/plain; charset="utf-8"

Dear all,

I hope this email finds you well.
I am interested in analysing longitudinal studies where a particular

group

of individuals (diagnosed at time 1) is followed across time and then

have

their skills measured at some later date (follow-up - time 2). I am not
interested in estimating the difference in skills between time points,

but

instead, I want to determine which factors measured at time 1 (e.g.,
gender, age) predict their skills at time 2. Assuming the models would be
regressions where the outcome variable at time 2 is predicted by each
factor at time 1 independently, could we use cohen's f as the effect size
for the meta-analysis and then run a meta-regression to see which factors
explain the most variance and which combinations lead to more explanatory
power? (e.g., voc ~ gender + SES). If this is completely wrong, could you
please point me to a study that has examined similar questions?

The dataset I am imagining would look something like this:

Study | Moderator | cohen's f | Outcome

S1 | gender | .23  | voc

S1 | SES | .12 | voc

S2| gender | .02 | voc


Thank you!

      [[alternative HTML version deleted]]




------------------------------

Subject: Digest Footer