I have two continuous independent variables and a binary dependent variable, and I want to use regression by means of polyserial correlation, but I can't find any information on how to do this in R. I know the alternative is to use logistic regresssion, but the problem that I'm trying to solve necessitates that I use polyserial correlation. I'm particularly concerned with getting the residuals of the regression. Any help would be much appreciated. If you know of another software that can perform this regression, I would also like to hear about it.

  • $\begingroup$ Why can't you use logistic regression? $\endgroup$ – Peter Flom - Reinstate Monica Mar 20 '12 at 11:06
  • $\begingroup$ Rseek.org tells me that the polycor package computes polychoric and polyserial correlations. Is that any help? $\endgroup$ – onestop Mar 20 '12 at 11:45
  • $\begingroup$ @PeterFlom I don't care about the coefficients. I care about the residuals, which I then plan to use in a SEM using package 'sem' in R, which relies on polychoric/polyserial correlations for categorical variables. $\endgroup$ – Joaquin Aguilar Mar 23 '12 at 2:52
  • $\begingroup$ @onestop Thank you, but unfortunately no. I'm already aware of this information. $\endgroup$ – Joaquin Aguilar Mar 23 '12 at 2:53

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.