Cluster Analysis of Ordinal Categorical data in JASP

Sport_biz_stats

The machine learning function in JASP is great, but I am working with a data set that involves only categorical data sets which are not appropriate for the current cluster methods given that the average for ordinal categorical data does not make sense. I am aware that I could map numerical values to my categories given they are ordered and use other methods (e.g., K-means or density). What would be the limitations of this approach (any further reading appreciated) and I wonder if K-Modes could/would be added from R to JASP to handle such data?

patc3

 I am aware that I could map numerical values to my categories given they are ordered and use other methods (e.g., K-means or density). What would be the limitations of this approach (any further reading appreciated) 

as this is a statistical question you would do well to either search on Google and/or ask on a statistics forum (Cross-Validated in particular, or even Reddit). That said, depending on your field (and the number of distinct categories your ordinal variables have), I think many people resort to treating their ordinal variables as numeric and do k-means.

I wonder if K-Modes could/would be added from R to JASP to handle such data?

That could be a good idea (I myself use Gower distances from the cluster package in R when I have non-numeric data for clustering). If you want to suggest a new feature you can do so here (you need to sign up for a GitHub account): https://github.com/jasp-stats/jasp-issues/issues/new/choose

TarandeepKang

Hi @Sport_biz_stats and @patc3, I'm not one of the development team but just another enthusiastic user, like the both of you. In any case, I saw this discussion and thought that this feature would be useful so I created the feature request. See here:

[Feature Request]: extend clustering algorithms to take account of categorical data and mixed data · Issue #2822 · jasp-stats/jasp-issues (github.com)

If you think it's missing anything, feel free to contribute! :-)

Best wishes,

Tarandeep

patc3

@TarandeepKang Cool! Great idea, I personally would love to see this arrive in JASP