[R-meta] Efficiency of coding scheme for meta-analysis
rnorouz|@n @end|ng |rom gm@||@com
Tue Nov 12 18:09:54 CET 2019
In view of the problems that categorical moderators have (e.g., categories
that apply only to one study, or even one group of a study) in
meta-analytic datasets which sometimes lead to some packages throw in
warnings or crash, I've have come with "*3 bare-minimum rules*" for the
efficiency of the coding schemes for categorical moderators.
I appreciate your input to expand, modify, add to or provide references on
any or all of these principles?
*Here are the 3 rules:*
(1) An efficient coding sheet is one on which *at least* two studies
receive a similar category on a moderator (e.g., two studies among all
studies for which moderator "X" is coded ex: "4").
(2) An efficient coding sheet is one on which *at least* two study groups
receive a similar category on a moderator (e.g., two groups for which
moderator "Y" is coded ex: "5").
(3) An efficient coding sheet is one on which there is no moderator with
more than 5 categories.
Postdoctoral Research Associate | Lecturer
Second Language Acquisition & Research Methods, Ph.D.
College of Education & Human Development
Dep. of Teaching, Learning & Culture | Texas A&M University
College Station, TX 77843
Email: rnorouzian using tamu.edu
*Future L2 researchers will be challenged not only on the basis of their
substantive questions, but also on how they manage to answer those
questions in a methodical manner.*
[[alternative HTML version deleted]]
More information about the R-sig-meta-analysis