On Tue, 18 Nov 2003, Jeff D. Hamann wrote:
> Can someone tell me why this is since I can't seem to find an
explination in
> the docs or FAQ for this.
>
> Since there are no "BM" in the mtrs data.frame, but only in the
"parent"
> data.frame (trees), I'm assuming the subset data.frame is still
associated
> with the original. Is that correct and how would I create a completly
> separate data.frame since I'll be using by, lapply and such on the
subset
> data.frame using "sp" which currently causes by to choke...
>
> > mtrs <- trees[trees$m == 1,]
> > mtrs
> plot tree sp m dbh tht bfvol cfvol
> 1 1 1 DF 1 44 185 3825.79 622.97
> 2 1 2 DF 1 38 188 3166.69 504.12
> 26 3 1 DF 1 42 185 3568.80 578.09
> 51 5 4 DF 1 30 189 2071.36 348.28
> 67 7 1 DF 1 33 168 2036.30 342.90
> 70 7 4 DF 1 45 180 3665.82 621.90
> 71 7 5 DF 1 30 160 1530.88 274.69
> 81 9 1 DF 1 44 175 3398.52 576.79
> 82 9 2 DF 1 68 205 9516.82 1461.21
> 95 11 1 DF 1 30 170 1775.60 299.54
> 96 11 2 DF 1 44 197 4249.98 680.64
> 97 11 3 DF 1 44 187 3914.88 632.42
> 107 11 13 WH 1 44 175 3398.52 576.79
> 121 13 2 WF 1 48 205 5241.14 827.57
> 126 13 7 DF 1 40 182 3188.77 522.43
> 138 15 2 DF 1 69 182 7852.22 1285.42
> 143 15 7 DF 1 28 142 1146.77 207.67
> 159 17 9 DF 1 12 88 99.13 25.43
> 177 19 7 DF 1 37 162 2309.06 391.48
> > attributes( mtrs$sp )
> $class
> [1] "factor"
>
> $levels
> [1] "BM" "DF" "WF" "WH"
>
A factor remembers all its levels unless you tell to drop them as in:
mtrs <- trees[trees$m == 1, , drop = TRUE]
Uwe Ligges