3
$\begingroup$

In SPSS, if I use the hierarchical clustering procedure, I have the ability to cluster both variables and cases using a variety of methods and distance measures. For this task, I would like to use R to cluster my variables.

For context, my data come from a survey and the respondents were able to select multiple items from a block of options. In my datafile the data are coded as 0,1.

I am trying to learn R so any help you can provide will be greatly appreciated.

Improve this question
$\endgroup$
3
  • $\begingroup$ How many respondents and items do you have? Can we assume there are different blocks of items, or do you put all items into the same box, so to speak? (This can help to figure out the best clustering solution.) $\endgroup$
    – chl
    Commented Mar 31, 2011 at 21:18
  • $\begingroup$ I'm not sure I understand how your dataset looks like. But you may want to transpose your data matrix (turning variables into cases and cases into variables) and run clustering. Would you find it useful to see how to do this in R? $\endgroup$
    – GaBorgulya
    Commented Mar 31, 2011 at 21:45
  • $\begingroup$ Does this relate to biclustering? That's a method for clustering both variables and cases simultaneously. See also this previous Q. $\endgroup$
    – onestop
    Commented Mar 31, 2011 at 22:01

1 Answer 1

Reset to default
2
$\begingroup$

Hi Take a look at the varclus function in the Frank Harrel's Hmisc package. 

require(Hmisc)
? varclus
Improve this answer
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.