# [R] simplify code for dummy coding of factors

Richard M. Heiberger rmh at temple.edu
Wed Dec 31 04:16:27 CET 2014

```It can handle one more simplification step

# Rich Heiberger, removing intercept, including haireye data
haireye.df <- cbind(
as.data.frame(haireye),
model.matrix(Freq ~ Hair + Eye, data=haireye,
contrasts.arg=list(Hair=diag(4), Eye=diag(4)))[,-1]
)
haireye.df

On Tue, Dec 30, 2014 at 9:15 PM, Michael Friendly <friendly at yorku.ca> wrote:
> On 12/30/14 7:16 PM, Richard M. Heiberger wrote:
>>
>> I like this very simple version.  Note that you don't need
>> as.data.frame().
>>
>> model.matrix(Freq ~ Hair + Eye, data=haireye,
>> contrasts.arg=list(Hair=diag(4), Eye=diag(4)))
>
> Thanks to all who replied to this question.  model.matrix() was what I had
> missed, but to
> get the result in the form I wanted, required a bit  more work. These
> solutions give me
>
> haireye <- margin.table(HairEyeColor, 1:2)
>
> # Jeff Newmiller, solution 2, corrected
>
> haireye.df <- as.data.frame(haireye)
> dummykeys <- data.frame( h = factor( as.integer( haireye.df\$Hair ) )
>                        , e = factor( as.integer( haireye.df\$Eye ) ) )
> dummy.hair <- as.data.frame( model.matrix( ~ h - 1, data=dummykeys ))
> dummy.eye <- as.data.frame( model.matrix( ~ e - 1, data=dummykeys ))
> haireye.df <- data.frame( haireye.df, dummy.hair, dummy.eye )
>
>
> # Rich Heiberger, removing intercept, including haireye data
> haireye.df <- as.data.frame(haireye)
> haireye.df <- cbind(
>     haireye.df,
>     model.matrix(Freq ~ Hair + Eye, data=haireye,
>         contrasts.arg=list(Hair=diag(4), Eye=diag(4)))[,-1]
>     )
> haireye.df
>
>
>
>
> --
> Michael Friendly     Email: friendly AT yorku DOT ca
> Professor, Psychology Dept. & Chair, Quantitative Methods
> York University      Voice: 416 736-2100 x66249 Fax: 416 736-5814
> 4700 Keele Street    Web:   http://www.datavis.ca
> Toronto, ONT  M3J 1P3 CANADA
>

```