Jesse Gervais
2013-Nov-28 18:23 UTC
[R] polychoric correlation with multiple imputations, a Strate and a Weight
Hi there, I'm generally more a Stata user than a R user, but I need to computed something, and I am not able to do it with Stata 13. So, here I am! I have a database that has multiple imputations (imputations are already done) with a complex sample design (Strate and Weight). Is it possible, in R, to run polychoric correlation with multiple imputation, a Strate and a Weight ? For the moment, I have been able to do some statistical analysis with Strate and Weight using the Survey package. I also know that there is a package named Amelia II that can handle multiple imputations, and polycor that can compute polychoric correlations. The problem is that I don't know how to "merge" all those packages together, if it is possible. So: 1- It is possible to do so ? 2- IF yes, I'm not sure to know how to merge all this together. Any ideas? I don't need any syntax (at least for the moment). I just need a hint of where to start (is there a package that can handle it, is there somebody who wrote paper(s) about it, etc.); I am completely stuck. Thank you! [[alternative HTML version deleted]]
Maybe Matching Threads
- Correlation when one variable has zero variance (polychoric?)
- polychoric correlation: issue with coefficient sign
- Polychoric and tetrachoric correlation
- Error messages when attempting to calculate polychoric correlation matrices using the psych package
- Polychoric Principal Component Analysis (pPCA)