If it converts into a factor although it appears to be numeric, then there is probably a string entry somewhere in this variable, which causes R to convert the whole variable into a factored variable (which results in dummies in the regression). You will want to check whether there are any excess delimiters (spaces or what not) in the observations that you have added.
Say x is your variable. Do is(x) before and after adding the 1000 observations. If x is numeric before and factor after adding the observations, then it tells you that the above mentioned problem exists in the added observations. Best, Daniel ------------------------- cuncta stricte discussurus ------------------------- -----Ursprüngliche Nachricht----- Von: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] Im Auftrag von Chris Poliquin Gesendet: Saturday, December 13, 2008 2:23 AM An: r-help@r-project.org Betreff: [R] Problems with large data frames? Hi, I recently added another 1,000 observations to a data frame I have for a total of 10,861 observations. Observations are by row and the frame has 15 columns. R now keeps converting certain columns to factors whenever I run regressions and can't seem to figure out the correct scaling for the axes of plots on its own. Are these problems the result of the size of my data frame or something else? I can't see anything wrong with the data itself. Any ideas what might be causing these recent problems and any way to force R not to convert variables to factors? - Chris ______________________________________________ R-help@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. ______________________________________________ R-help@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.
