[R] WEIGHT CASES by Frequency VECTOR

[Peter Dalgaard]

sharon Wandia wrote:
> Does anyone know how to weight cases in a data frame using a frequency
> vector?
> I'm trying to run tabulations on R , on a data set that first needs to have
> weighted cases before i run the tabulations.
> 
> In SPSS & SAS its quite simple, but i'm unable to do it in R.


xtabs() allows a count on the LHS of its formula argument, if you're
looking at contingency tables. In other cases, an expedient, if
inefficient, trick is simply to expand your data using rep(), i.e.
something like

mydata.x <- mydata[rep(1:nrow(mydata),mydata$count),]

(Re. SPSS & SAS: what is not quite as simple is to see whether weights
are frequency weights or variance weights.)

-- 
   O__  ---- Peter Dalgaard             Øster Farimagsgade 5, Entr.B
  c/ /'_ --- Dept. of Biostatistics     PO Box 2099, 1014 Cph. K
 (*) \(*) -- University of Copenhagen   Denmark      Ph:  (+45) 35327918
~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk)              FAX: (+45) 35327907