使用GGPLOT2

Question

序因子濃度直我有一個簡單的數據集，看起來像這樣：

school  score  grade subject 
Your school 83.27  1  English 
All schools 113.60  6  English 

等級爲9級序的因素，學校是2級的一個因素。

'data.frame': 12762 obs. of 4 variables: 
$ school : Factor w/ 2 levels "Your school",..: 1 2 2 2 2 2 2 2 2 2 ... 
$ score : num 83.3 113.6 109.2 117.4 100.3 ... 
$ grade : Ord.factor w/ 9 levels "1"<"2"<"3"<"4"<..: 1 6 6 7 4 3 6 1 6 6 ... 
$ subject: chr "English" "English" "English" "English" ... 

我想用ggplot2使用密度函數繪製一個多面直方圖。如果我嘗試沒有密度我得到這個：

p <- ggplot(ss, aes(x=grade)) 
p <- p + geom_histogram() 
p <- p + facet_wrap(~school) 

當我添加密度正常化y軸我得到這個：

p <- ggplot(ss, aes(x=grade)) 
p <- p + geom_histogram(aes(y=..density..)) 
p <- p + facet_wrap(~school) 

什麼我錯過了嗎？

Answer 1

R無法計算一個因子的密度，甚至排序一個因子。你最好的選擇是將grade轉換爲數字（例如使用as.numeric(as.character(x))）。

xy <- data.frame(school = sample(c("your", "all"), size = 100, replace = TRUE), 
        grade = sample(1:10, size = 100, replace = TRUE)) 
xy$grade.factor <- factor(xy$grade, ordered = TRUE) 

library(ggplot2) 

# doesn't work for factors 
ggplot(xy, aes(x = grade.factor)) + 
    theme_bw() + 
    geom_histogram(aes(y = ..density..)) + 
    facet_wrap(~ school) 

# works for integers/numeric 
ggplot(xy, aes(x = grade)) + 
    theme_bw() + 
    geom_histogram(aes(y = ..density..)) + 
    facet_wrap(~ school)