Can anyone point me to examples with R code where bwplot in lattice is used to order the boxes in ascending order? I have found the following discussion and it partly works. But, I have a conditioning variable, so my example is more like bwplot(var1 ~ var2|condition, dat) Th example in the discussion below works only when there is not a conditioning variable as far as I can tell. I can tweak the example below to work, but then I get some ugly labels in the lattice plot. It seems index.cond is supposed to help me solve this, but I cannot find good examples showing its use. Thanks Harold http://r.789695.n4.nabble.com/bwplot-reorder-factor-on-y-axis-td790903.html
bwplot in ascending order
4 messages · Doran, Harold, Mark Difford, Uwe Ligges
On May 01 (2011) Harold Doran wrote:
Can anyone point me to examples with R code where bwplot in lattice is used to order the boxes in ascending order?
You don't give an example and what you want is not entirely clear. Presumably you want ordering by the median (boxplot, and based on the example you point to, where the median is mentioned as an _example_). Is this what you want? ## bwplot(var1 ~ var2|condition, dat, index.cond = function(x, y) reorder(y, x, median)) ## if x is numeric bwplot(var1 ~ var2|condition, dat, index.cond = function(x, y) reorder(x, y, median)) ## if y is numeric Regards, Mark. -- View this message in context: http://r.789695.n4.nabble.com/bwplot-in-ascending-order-tp3488557p3489544.html Sent from the R help mailing list archive at Nabble.com.
On 01.05.2011 22:52, Doran, Harold wrote:
Can anyone point me to examples with R code where bwplot in lattice is used to order the boxes in ascending order? I have found the following discussion and it partly works. But, I have a conditioning variable, so my example is more like bwplot(var1 ~ var2|condition, dat)
I guess you are looking for something along bwplot(var1 ~ var2 | reorder(condition, var2, median), dat) Uwe Ligges
Th example in the discussion below works only when there is not a conditioning variable as far as I can tell. I can tweak the example below to work, but then I get some ugly labels in the lattice plot. It seems index.cond is supposed to help me solve this, but I cannot find good examples showing its use. Thanks Harold http://r.789695.n4.nabble.com/bwplot-reorder-factor-on-y-axis-td790903.html
______________________________________________ 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.
Doesn't seem to work. My data structure is below (I will send data to anyone off-list who could offer support). The following code below does work, but since I concatenate Region and Gender, the labels on the lattice are ugly. dat$test <- factor(paste(dat$Region, dat$Gender, sep='_')) bymedian <- with(dat, reorder(test, finalRank, median)) bwplot(reorder(test, finalRank, median) ~ finalRank|Gender, dat, subset = Region !="", scale='free', xlab = 'Total Score', ylab = 'Region', )
str(dat)
'data.frame': 58921 obs. of 16 variables: $ Athlete : int 13 13 13 13 13 14 14 15 15 15 ... $ Workout : Factor w/ 6 levels "11.1","11.2",..: 1 2 3 4 5 1 2 1 1 2 ... $ Result : int 309 375 46 100 300 158 232 353 359 479 ... $ Valid : Factor w/ 6 levels "bogus","invalid",..: 5 5 5 5 5 5 5 5 5 5 ... $ Gender : Factor w/ 2 levels "female","male": 2 2 2 2 2 2 2 2 2 2 ... $ Height.cm.: num 196 196 196 196 196 ... $ Weight.kg.: num 97.7 97.7 97.7 97.7 97.7 ... $ Age : int 29 29 29 29 29 42 42 24 24 24 ... $ Region : Factor w/ 18 levels "","Africa","Asia",..: 16 16 16 16 16 13 13 18 18 18 ... $ AgeCut : num 2 2 2 2 2 4 4 2 2 2 ... $ Height.met: num 1.96 1.96 1.96 1.96 1.96 ... $ spVar : chr "11.1_male" "11.2_male" "11.3_male" "11.4_male" ... $ Rank : int 1567 2253 2050 1651 1462 8155 7624 322 208 206 ... $ totalRank : int [1:58921(1d)] 8983 8983 8983 8983 8983 15779 15779 1252 1252 1252 ... ..- attr(*, "dimnames")=List of 1 .. ..$ : chr "13" "13" "13" "13" ... $ finalRank : int 1274 1274 1274 1274 1274 2643 2643 81 81 81 ... $ totalScore: int [1:58921(1d)] 1130 1130 1130 1130 1130 390 390 1768 1768 1768 ... ..- attr(*, "dimnames")=List of 1 .. ..$ : chr "13" "13" "13" "13" ...
-----Original Message----- From: Uwe Ligges [mailto:ligges at statistik.tu-dortmund.de] Sent: Monday, May 02, 2011 4:58 AM To: Doran, Harold Cc: r-help at r-project.org Subject: Re: [R] bwplot in ascending order On 01.05.2011 22:52, Doran, Harold wrote:
Can anyone point me to examples with R code where bwplot in lattice is used
to order the boxes in ascending order? I have found the following discussion and it partly works. But, I have a conditioning variable, so my example is more like
bwplot(var1 ~ var2|condition, dat)
I guess you are looking for something along bwplot(var1 ~ var2 | reorder(condition, var2, median), dat) Uwe Ligges
Th example in the discussion below works only when there is not a
conditioning variable as far as I can tell. I can tweak the example below to work, but then I get some ugly labels in the lattice plot. It seems index.cond is supposed to help me solve this, but I cannot find good examples showing its use.
______________________________________________ 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.