[R] Function similar to combine.levels in Hmisc package

[Frank E Harrell Jr]

On 07/09/2010 05:33 PM, Saeed Abu Nimeh wrote:
> Is there a function similar to combine.levels ( in the Hmisc package)
> that combines the levels of factors, but not based on their frequency.
> Alternatively, I am looking into using the significance of the dummy
> variables of factors based on their Pr(>|t|) value using the linear
> model, then deleting the non-significant levels. Any other
> suggestions?
> Thanks,
> Saeed

That is not a valid statistical procedure.  You will not have the 
correct d.f. in the final F test.

Frank

-- 
Frank E Harrell Jr   Professor and Chairman        School of Medicine
                      Department of Biostatistics   Vanderbilt University