I'm working with R and want to run a correspondance analysis on a dataset containing, among the others, the following factors:

city district: 27 levels, each corresponding to a given district

objective 1, objective 2, ... - factors with two levels (yes/no) corresponding to whether or not a person has chosen a given objective as important to him.

The question is: Does it matter if I transform the single variable city district into 27 binary variables? Would it impact the outcome of a correspondence analysis, including how good the data variance is described by each of the principal components?


migrated from stackoverflow.com Jan 21 at 13:27

This question came from our site for professional and enthusiast programmers.

  • $\begingroup$ You may want to take a look at this article: utdallas.edu/~herve/Abdi-MCA2007-pretty.pdf. R implementations of the MCA will transform themselves the factors into matrices of zeros and ones. $\endgroup$ – Vincent Guillemot Jan 21 at 13:15
  • $\begingroup$ Thatk you for your comment! I'll take a look. $\endgroup$ – pw29 Jan 24 at 20:22

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.