change the order in geom_bar of the stacked variables

Alessandra Carioli

unread,

Jan 26, 2016, 2:00:05 PM1/26/16

to ggplot2

Dear ggplotters,

I am trying to change the stacking order of categories so that “Recent change in preferences” appears plotted last, after “observed”, “competing preferences” , “waiting time”, and “strict involuntary factors”

I have created an ordering variable, numV and added order in the aes but it does not work:

numV = factor( dt$Type, c(“Observed”, "Competing Preferences", ”Waiting time", "Strict Involuntary factors", "Recent change in Preferences"))

ggplot( subset( dt, Variable == “TFR1”), aes ( reorder( Country, rev( order )), fill = numV, order= Type, value))+

geom_bar( stat= "identity") +

scale_fill_manual(values=c( "brown", "lightblue", "purple”, "darkgreen", "green"),breaks=c("Observed", "Competing Preferences", "Waiting time", "Strict Involuntary factors", "Recent change in Preferences" ))+

coord_flip()+

geom_hline( yintercept= 1, col= "red", lty=2)

dt <- structure(list(Country = structure(c(1L, 2L, 8L, 11L, 6L, 10L,

3L, 4L, 5L, 7L, 9L, 12L, 1L, 2L, 8L, 11L, 6L, 10L, 3L, 4L, 5L,

7L, 9L, 12L, 1L, 2L, 8L, 11L, 6L, 10L, 3L, 4L, 5L, 7L, 9L, 12L,

1L, 2L, 8L, 11L, 6L, 10L, 3L, 4L, 5L, 7L, 9L, 12L, 1L, 2L, 8L,

11L, 6L, 10L, 3L, 4L, 5L, 7L, 9L, 12L), .Label = c("Austria",

"Belgium", "Bulgaria", "Czech Rep", "Estonia", "Finland", "Hungary",

"Italy", "Lithuania", "Norway", "Portugal", "Slovenia"), class = "factor"),

Type = structure(c(2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,

2L, 2L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 1L,

1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 5L, 5L, 5L, 5L,

5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,

4L, 4L, 4L, 4L, 4L), .Label = c("Competing Preferences",

"Observed", "Recent change in Preferences", "Strict Involuntary factors",

"Waiting time"), class = "factor"), Variable = structure(c(3L,

3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L,

3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L), .Label = c("PPR",

"TFR", "TFR1"), class = "factor"), value = c(0.915019618935318,

0.983361531957373, 0.914760769151969, 0.859972469668077,

0.844763634128951, 0.9304363613935, 0.871577019929315, 0.973252000207427,

0.751675755207165, 0.989007326062785, 0.883086301186579,

0.86635016841834, 0, 0.007408699, 0, 0, 0.02554987, 0.011588498,

0, 0, 0, 0.025129468, 0, 0, 0.069921961, 0, 0.046661336,

0.125630417, 0, 0, 0.121056141, 0.016275558, 0.226832447,

0, 0.108421811, 0.121703024, 0.008784752, 0.010205031, 0.019966701,

-0.000427248, 0.005958654, 0.005720579, 0.001671895, 0.006065908,

0.010108353, 0.003915961, 0.001104456, 6.96476e-05, 0.006273667,

0.006433437, 0.018611195, 0.014824362, 0.149277712, 0.063843059,

0.005694944, 0.004406534, 0.011383445, -0.018052755, 0.007387433,

0.01187716), order = c(1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L,

10L, 11L, 12L, 1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L, 10L, 11L,

12L, 1L, 2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L, 10L, 11L, 12L, 1L,

2L, 3L, 4L, 5L, 6L, 7L, 8L, 9L, 10L, 11L, 12L, 1L, 2L, 3L,

4L, 5L, 6L, 7L, 8L, 9L, 10L, 11L, 12L), numV = c(1L, 1L,

1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 5L, 5L, 5L, 5L, 5L,

5L, 5L, 5L, 5L, 5L, 5L, 5L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,

2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L,

3L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L)), .Names = c("Country",

"Type", "Variable", "value", "order", "numV"), row.names = c(25L,

26L, 27L, 28L, 29L, 30L, 31L, 32L, 33L, 34L, 35L, 36L, 61L, 62L,

63L, 64L, 65L, 66L, 67L, 68L, 69L, 70L, 71L, 72L, 97L, 98L, 99L,

100L, 101L, 102L, 103L, 104L, 105L, 106L, 107L, 108L, 133L, 134L,

135L, 136L, 137L, 138L, 139L, 140L, 141L, 142L, 143L, 144L, 169L,

170L, 171L, 172L, 173L, 174L, 175L, 176L, 177L, 178L, 179L, 180L

), class = "data.frame")

Gray, Wayne

unread,

Jan 26, 2016, 3:30:41 PM1/26/16

to Alessandra Carioli, ggplot2

Alessandra,

## REORDERING LEVELS

levels(e2eye$DIRTIME)

[1] "EARLY" "LATE" "MIDDLE"

e2eye$DIRTIME <- factor(e2eye$DIRTIME, levels=c("EARLY", "MIDDLE", "LATE"))

levels(e2eye$DIRTIME)

[1] "EARLY" "MIDDLE" “LATE"

Works well.

Wayne

--
--
You received this message because you are subscribed to the ggplot2 mailing list.
Please provide a reproducible example: https://github.com/hadley/devtools/wiki/Reproducibility

To post: email ggp...@googlegroups.com
To unsubscribe: email ggplot2+u...@googlegroups.com
More options: http://groups.google.com/group/ggplot2

---
You received this message because you are subscribed to the Google Groups "ggplot2" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ggplot2+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Gray, Wayne

unread,

Jan 27, 2016, 9:50:58 PM1/27/16

to Alessandra Carioli, ggplot2

Alessandra,

I give up!!

I spent some time on your problem — starting by simplifying it to eliminate everything but what was causing your problem:

dt$Type <- factor(dt$Type, levels=c("Observed", "Competing Preferences", "Waiting time", "Strict Involuntary factors", "Recent change in Preferences"))

ggplot(dt, aes(x=Country, y=value, fill = Type)) +

geom_bar(stat="identity”)

So — this eliminates the coord_flip(), as well as reversing the ordering of “Country”, it also removes your assignment of color and breaks. It is pretty much the simplest stacked-barchart of these data and I am still not able to control the order in which the stacks stack.

Perhaps someone else can help us both out.

Wayne

BTW: you have two values that are < 0. Row 40 and row 58. You probably want to eliminate those.

On 16-Jan-27, at 04:03 , Alessandra Carioli <alessand...@gmail.com> wrote:

Hi Wayne,

it does reorder the levels in the legend but not in the graph. Any ideas how to do it?

Ale

ggplot(subset(dt,Variable=="TFR1"),aes(reorder(Country,rev(order)),fill=Type,value))+

geom_bar(stat="identity") +

scale_fill_manual(values=c("lightblue", "brown", "green","darkgreen","purple"),breaks=c("Observed","Competing Preferences","Waiting time","Strict Involuntary factors","Recent change in Preferences"))+

coord_flip()

Ben

unread,

Jan 29, 2016, 6:17:57 AM1/29/16

to ggplot2, alessand...@gmail.com

First, let me echo Wayne's comment--please SIMPLIFY reproducible examples as much as possible.

I think this is a long-standing bug (https://github.com/hadley/ggplot2/issues/721), but it can be worked around:

# using arrange from dplyr:

ggplot(arrange(dt, Type), aes(x=Country, y=value, fill = Type)) + geom_bar(stat='identity')

# using base R:

ggplot(dt[order(dt$Type), ], aes(x=Country, y=value, fill = Type)) + geom_bar(stat='identity')

Ben

Reply all

Reply to author

Forward