Graphical Representation of large data

[varsha]

Hi,

I am a stats newbie. I have to show how 150 variables vary across 13 groups using a graphical representation. Using the usual graphical representations become too complex. I wanted to know if there is a way to represent this without coming up with too complex graphs.

Thanks,
Varsha.

[Mean Joe]

You could do principal components analysis. This tutorial explains it a bit, and also gives code in R.

Not sure how much of a newbie you are, or even what kind of variables you're working with.

[gianmarco]

Hi,
just to add that if you are dealing with categorical variables, you may use Correspondence Analysis. It allows you to display contingency table in visual form (i.e. scatterplot), and provides information to explore pattern(s) of associations in the dataset.

There are many previous posts in this Forum. You could try to make a search.
Regards

Gm

[bugman]

There are numerous options available to you, but at the end of the day it comes down to the type of data you have. PCA is fine under certain criteria, as is CA but we would need to know more about your data.

P