An embedded and charset-unspecified text was scrubbed... Name: not available URL: <https://stat.ethz.ch/pipermail/r-help/attachments/20111215/678ef47b/attachment.pl>
Reordering a numeric variable
2 messages · Abraham Mathew, Bert Gunter
I believe this needs to be posted on general statistical help list like stats.stackexchange.com as you appear to have some confusion about how linear regression works, especially with categorical variables. ?ordered might help you with R's way of dealing with the issues that I believe you're asking about, but it may not make much sense if I'm correct about your underlying misunderstanding. If I'm wrong, that and perhaps ?C will probably do it for you. -- Bert
On Thu, Dec 15, 2011 at 1:44 PM, Abraham Mathew <abraham at thisorthat.com> wrote:
I'm running a linear model in R using the car package. I have a variable education, which i have recoded and regrouped to my wishes. However, R seems to place each element of that variable in alphabetical order. When I am running the model, don't I need the model order from lowest to highest to make an inference that a one unit change in one variable produced a one unit change in another. levels(educ) educ2 = NA educ2[educ %in% levels(educ)[c(4,7)]] <- "HS or Some College" educ2[educ %in% levels(educ)[1:2]] <- "College Degree" educ2[educ %in% levels(educ)[c(3,5)]] <- "Advanced Degree" educ2[educ %in% levels(educ)[c(6,8)]] <- "Other" educ2 = factor(educ2) levels(educ2) The above code is how I regrouped the variable. How can I regroup it so that it's levels are from lowest to highest. What if they're numeric values" -- *Abraham Mathew Statistical Analyst This or That Media, Inc. abraham at thisorthat.com 720-648-0108 @abmathewks www.amathew.com * ? ? ? ?[[alternative HTML version deleted]]
______________________________________________ R-help at r-project.org mailing list https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code.
Bert Gunter Genentech Nonclinical Biostatistics Internal Contact Info: Phone: 467-7374 Website: http://pharmadevelopment.roche.com/index/pdb/pdb-functional-groups/pdb-biostatistics/pdb-ncb-home.htm