Coord_flip()在成组的条形图中更改组内条的顺序 [英] Coord_flip() changes ordering of bars within groups in grouped bar plot
问题描述
我创建了一个组条形图(img 1)，并使用coord_flip()移动了
将条形图分组到y轴.我注意到coord_flip()还重新排列了每个条在组中的显示方式.
I have created a group bar plot (img 1) and used coord_flip() to move the
groups bars to the y axis. I noticed that coord_flip() also reorders the way each bar is presented in the groups.
例如在img中，条形图来自A-D组.但是，在img 2中，条形图来自D-A组.如何使用coord_flip()保持每个组中条形的相同顺序?
E.g. In img 1 bars flow from Group A-D. However, in img 2 the bars flow from Group D-A. How can I keep the same ordering of the bars within each group using coord_flip()?
ggplot(all_Q, aes(x=qid, y=correct_per, fill=group)) +
geom_bar(stat="identity", position="dodge")
ggplot(all_Q, aes(x=qid, y=correct_per, fill=group)) +
geom_bar(stat="identity", position="dodge") +
scale_x_discrete(limits = as.character(16:1)) +
coord_flip()
工作示例(数据子集-问题8-11)
Working example (subset of data -- questions 8-11)
dput() output:
structure(list(group = structure(c(1L, 2L, 3L, 4L, 1L, 2L, 3L, 4L, 1L, 2L, 3L, 4L, 1L, 2L, 3L, 4L), .Label = c("A", "B", "C", "D"), class = "factor"), correct_per = c(90.4761904761905, 100, 100, 87.5, 83.3333333333333, 90.9090909090909, 84.6153846153846, 87.5, 80.9523809523809, 88.6363636363636, 100, 70.8333333333333, 63.4146341463415, 76.7441860465116, 76.9230769230769, 62.5), nr_correct = c(38L, 44L, 26L, 21L, 35L, 40L, 22L, 21L, 34L, 39L, 26L, 17L, 26L, 33L, 20L, 15L), nr_incorrect = c(4L, 0L, 0L, 3L, 7L, 4L, 4L, 3L, 8L, 5L, 0L, 7L, 15L, 10L, 6L, 9L), length = c(42L, 44L, 26L, 24L, 42L, 44L, 26L, 24L, 42L, 44L, 26L, 24L, 41L, 43L, 26L, 24L), qid = c("8", "8", "8", "8", "9", "9", "9", "9", "10", "10", "10", "10", "11", "11", "11", "11")), .Names = c("group", "correct_per", "nr_correct", "nr_incorrect", "length", "qid"), row.names = c(NA, -16L), class = c("tbl_df", "tbl", "data.frame"))
save to file
all_Q <- dget(filename)
ggplot(all_Q, aes(x=qid, y=correct_per, fill=group)) +
geom_bar(stat="identity", position="dodge") +
scale_x_discrete(limits = as.character(11:8)) +
coord_flip()
推荐答案
一种"hacky"方法是将position_dodge宽度设置为负数. .9是默认设置，因此-.9会创建相同的外观，但相反.
One "hacky" way to do this is to set the position_dodge width to be a negative number. .9 is the default, so -.9 would create the same appearance but in reverse.
library(ggplot2)
all_Q <- structure(list(group = structure(c(1L, 2L, 3L, 4L, 1L, 2L, 3L, 4L, 1L, 2L, 3L, 4L, 1L, 2L, 3L, 4L), .Label = c("A", "B", "C", "D"), class = "factor"), correct_per = c(90.4761904761905, 100, 100,
87.5, 83.3333333333333, 90.9090909090909, 84.6153846153846, 87.5, 80.9523809523809, 88.6363636363636, 100, 70.8333333333333, 63.4146341463415, 76.7441860465116, 76.9230769230769, 62.5), nr_correct = c(38L, 44L, 26L, 21L, 35L, 40L, 22L, 21L, 34L, 39L, 26L, 17L, 26L, 33L, 20L, 15L), nr_incorrect = c(4L, 0L, 0L, 3L, 7L, 4L, 4L, 3L, 8L, 5L, 0L, 7L, 15L, 10L, 6L, 9L), length = c(42L, 44L, 26L, 24L, 42L, 44L, 26L, 24L, 42L, 44L, 26L, 24L, 41L, 43L, 26L, 24L), qid = c("8", "8", "8", "8", "9", "9", "9", "9", "10", "10", "10", "10", "11", "11", "11", "11")), .Names = c("group", "correct_per", "nr_correct", "nr_incorrect", "length", "qid"), row.names = c(NA,
-16L), class = c("tbl_df", "tbl", "data.frame"))
ggplot(all_Q, aes(x = qid, y = correct_per, fill = group)) +
geom_bar(stat = "identity", position = position_dodge(-.9)) +
scale_x_discrete(limits = as.character(11:8)) +
coord_flip()
或者，您可以重新排列因子的水平，然后反转图例.
Alterantively, you could reorder the levels of the factor, then reverse the legend.
all_Q$group <- factor(all_Q$group, levels = rev(levels(all_Q$group)))
ggplot(all_Q, aes(x = qid, y = correct_per, fill = group)) +
geom_bar(stat = "identity", position = "dodge") +
scale_x_discrete(limits = as.character(11:8)) +
coord_flip() +
guides(fill = guide_legend(reverse = TRUE))
结果:
Result:
这篇关于Coord_flip()在成组的条形图中更改组内条的顺序的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！
