How to avoid recycling of colors in barplot to achieve different colors within each group?

Question

This is closely related to another question I just asked.

This is my data:

y <- structure(
  c(0.5619, 0.4381, 0.7587, 0.2413, 0.8764, 0.1236, 
    0.9019, 0.0981, 0.9481, 0.0519, 0.99, 0.01),
  .Dim = c(2L, 6L), .Dimnames = list(c("FALSE", "TRUE"), NULL)
)
y
#         [,1]   [,2]   [,3]   [,4]   [,5] [,6]
# FALSE 0.5619 0.7587 0.8764 0.9019 0.9481 0.99
# TRUE  0.4381 0.2413 0.1236 0.0981 0.0519 0.01

The original plot with same colors within each group (blue and red):

barplot(
  y, horiz = TRUE, col = c("blue", "red"),
  names.arg = c("Overall", paste("Flag", 5:1)), las = 1L,
  cex.names = 0.6,
  main = "Proportion Dropped Given Each Sample Restriction"
)

I want to change the red right-hand bars for each group, and instead have different colors within each group, something like:

I therefore created a new col vector with one color for each bar segment:

barplot(
  y, horiz =TRUE,
  col = c(
    "blue", "gold",
    "blue", "springgreen",
    "blue", "orange",
    "blue", "red",
    "blue", "white"
  ),
  names.arg = c("Overall", paste("Flag", 5:1)), las = 1L,
  cex.names = 0.6,
  main = "Proportion Dropped Given Each Sample Restriction"
)

However, only the first two colors in col (blue and gold) are used and they are recycled 6 times:

Is there any way to get the output I'm looking for?

Not an answer to your question, but once noticed a similar recycling of colours in `stripchart` (see [**here**](http://stackoverflow.com/questions/22920381/different-coloring-of-groups-in-r-plot/22923960#22923960) "Only as many colours that are needed to for the points in one row (for `col` colours) or column (for `bg` colours) are picked from the colour vector, then they are recycled.". For `barplot` it seems like "Only as many colours that are needed to for the _bars_ in one _column_ (for `col` colours) are picked from the colour vector, then they are recycled." — Henrik, Aug 05 '15 at 19:28
Border colors are also recycled. A tiny example: `m <- matrix(1:4, nrow = 2)`; `cols <- c("blue", "red", "green", "yellow")`; `border <- c("red", "blue", "yellow", "green")`; `par(lwd = 6)`; `barplot(m, col = cols, border = border)`. I have seen work-arounds for `base`, but I am not able to find them right now. You may consider `ggplot` and `scale_fill_manual`. — Henrik, Aug 05 '15 at 19:31
Relevant R-help posts [**here**](https://stat.ethz.ch/pipermail/r-help/2011-July/284640.html) and [**here**](https://stat.ethz.ch/pipermail/r-help/2007-March/126848.html). — Henrik, Aug 05 '15 at 19:33

score 2 · Accepted Answer · edited May 23 '17 at 11:44

2

Perhaps by tricking barplot into thinking there are more categories. This is tremendously ugly, but it seems to get the job done:

ymod <- cbind(c(y[,1], 0, 0,0,0,0,0,0,0,0,0),
              c(0, 0, y[,2],0,0,0,0,0,0,0,0),
              c(0, 0,0,0, y[,3],0,0,0,0,0,0),
              c(0, 0,0,0,0,0, y[,4],0,0,0,0),
              c(0, 0,0,0, 0,0,0,0,y[,5],0,0),
              c(0, 0,0,0, 0,0,0,0,0,0,y[,6]))

barplot(ymod, horiz=T,
        col=c(rbind(rep("blue",6),c("white","red","purple",
                                    "orange","springgreen","gold"))))

Here and here are some references that I used to achieve this result.

edited May 23 '17 at 11:44

Community

1
1

answered Aug 05 '15 at 20:44

erasmortg

3,246
1
17
34

2

Well, we can make this much nicer with `magic::adiag`, as inspired by [this](http://stackoverflow.com/questions/17495841/block-diagonal-binding-of-matrices): `ymod<-Reduce(magic::adiag,lapply(split(t(y),1:ncol(y)),as.matrix))` – MichaelChirico Aug 05 '15 at 21:55
Yes, beautiful! I struggled with transposing `y` and attempting to fit a pattern that would make it much more readable. This one is it! – erasmortg Aug 05 '15 at 22:44

How to avoid recycling of colors in barplot to achieve different colors within each group?

1 Answers1

Linked