For example ,
I would like to integrate all significant variables into one model to see if tau index is significantly decreased.
But I realized that , from the example from Cheung;
" http://courses.nus.edu.sg/course/psycwlm/internet/metaSEM/index.html"
I would like to put all predictors "Year, Discpline, Country.." into one model as final.
I realized that 'cbind()' function is limited to variable like "Discipline".
If i try to add "Year" to the equation, a error message appears.
> summary(Model5 <- meta3(y=fisherz, v=v, cluster=id, x=cbind(weeks, timepoint,(")dissertation("),(")peerreview(")), data=KORdb, intercept.constraints=0, model.name="Model5"))
Weeks, timepoint are continuous whereas dissertation and peerreview are processed as dicotomous.
How can i integrate variables with different characteristics into one model??
Dear James,
Dummy variables may be used to represent categorical variables. There are examples at http://courses.nus.edu.sg/course/psycwlm/internet/metaSEM/3level.html#sec-3
Mike
It didn't work last night but i re-coded every dummy variables once again to make sure each variable
is correctly specified. then it works :).