tabulate: switching columns and rows

Ricardo Rodríguez · 2006-12-31T18:44:04Z

>>> hadley wickham 31/12/2006 19:33 >>> >Hi Ricardo, >You might want to have a look at the reshape package, >http://had.co.nz/reshape, which provides a more general and flexible >framework for reshaping data in R. > >The version of warpbreaks I have doesn't have the replicate variable, >so unfortunately I can't provide you with example code for this case. >However, you should be able to figure it out for your data using the >documentation available on the website. Than

2 messages · Ricardo Rodríguez, Hadley Wickham

Original

Ricardo Rodríguez

Sun, Dec 31, 2006 10:44 AM #

Thanks, Hadley,
 
My fault: I've not provided the line concerning replicate variable. It is included in ?xtabs:

warpbreaks$replicate <- rep(1:9, len = 54)

It will be great to have an example with these given data. Anyway, I will have a look to reshape at the given site. Thanks a lot!

Cheers,

Ricardo




--
Ricardo Rodr?guez
Your XEN ICT Team

Hadley Wickham

Sun, Dec 31, 2006 10:58 AM #

install.packages("reshape")
library(reshape)

wm <- melt(warpbreaks, m="breaks")
cast(wm, wool ~ tension + replicate)

# Or as a 3d array
cast(wm, wool ~ tension ~ replicate)

# Or aggregate over replicates:
cast(wm, wool ~ tension, mean)
cast(wm, wool ~ tension, length)

Regards,

Hadley