1070423EX

EX1

把資料顯示為橫向資料密度長條圖，左半部及右半部面積各為1，易於辨識所佔百分比

library(pacman)
p_load(tidyverse)

dta <- read.table("hs0.txt", h = T) %>%
 mutate(female = factor(female, levels(female), 
                 labels = c("Female", "Male")),
        race = factor(race, levels(race), 
               labels = c("Black", "Asian", "Hispanic", 
                          "White")),
        ses = ordered(ses, levels = c("low", "middle", 
                      "high"),
              labels = c("Low", "Middle", "High"))) %>%
 mutate(race = reorder(race, math, median))

str(dta)

## 'data.frame':    200 obs. of  11 variables:
##  $ id     : int  70 121 86 141 172 113 50 11 84 48 ...
##  $ female : Factor w/ 2 levels "Female","Male": 2 1 2 2 2 2 2 2 2 2 ...
##  $ race   : Factor w/ 4 levels "Black","Hispanic",..: 3 3 3 3 3 3 1 2 3 1 ...
##   ..- attr(*, "scores")= num [1:4(1d)] 45 61 47 54
##   .. ..- attr(*, "dimnames")=List of 1
##   .. .. ..$ : chr  "Black" "Asian" "Hispanic" "White"
##  $ ses    : Ord.factor w/ 3 levels "Low"<"Middle"<..: 1 2 3 3 2 2 2 2 2 2 ...
##  $ schtyp : Factor w/ 2 levels "private","public": 2 2 2 2 2 2 2 2 2 2 ...
##  $ prog   : Factor w/ 3 levels "academic","general",..: 2 3 2 3 1 1 2 1 2 1 ...
##  $ read   : int  57 68 44 63 47 44 50 34 63 57 ...
##  $ write  : int  52 59 33 44 52 52 59 46 57 55 ...
##  $ math   : int  41 53 54 47 57 51 42 45 54 52 ...
##  $ science: int  47 63 58 53 53 63 53 39 58 NA ...
##  $ socst  : int  57 61 31 56 61 61 61 36 51 51 ...

bw <- with(dta, IQR(math)/(length(math)^(1/3)))
ggplot() +
 stat_bin(data = subset(dta, female=="Male"), binwidth = bw,
     aes(math, color = "Male", fill = "Male", y = - ..density.. )) +
 stat_bin(data = subset(dta, female == "Female"), binwidth = bw,
     aes(math, color = "Female", fill = "Female", y = ..density.. )) +
 scale_color_manual(values = c("black", "black"),
      guide = guide_legend(title = NULL, direction = "horizontal",
       title.position = "top", reverse = TRUE,
       label.position = "bottom", label.hjust = .5, label.vjust = .5,
       label.theme = element_text(angle = 90) ) ) +
 scale_fill_manual(values = c("White", "gray80"),
      guide = guide_legend(title = NULL, reverse = TRUE,
       direction = "horizontal", title.position = "top",
       label.position = "bottom", label.hjust = .5, label.vjust = .5,
       label.theme = element_text(angle = 90))) +
 scale_x_continuous(limits = c(30, 80), breaks=seq(30, 80, by = 5)) +
 labs(x = "Mathematic score", y = "Density") +
 coord_flip() +
 theme_bw() +
 theme(legend.position=c(.9, .85))

EX2

這是老師提供講義P39的例子(全部丟到迴歸方程式lm裡,再作圖)

pacman::p_load(tidyverse, gridExtra, ggExtra,ggfortify, GGally, ggthemes, ggrepel,
gganimate, ggeffects)

## Installing package into 'C:/Users/123/Documents/R/win-library/3.4'
## (as 'lib' is unspecified)

## Warning: package 'gganimate' is not available (for R version 3.4.3)

## Bioconductor version 3.6 (BiocInstaller 1.28.0), ?biocLite for help

## Warning in p_install(package, character.only = TRUE, ...):

## Warning in library(package, lib.loc = lib.loc, character.only = TRUE,
## logical.return = TRUE, : there is no package called 'gganimate'

## Warning in pacman::p_load(tidyverse, gridExtra, ggExtra, ggfortify, GGally, : Failed to install/load:
## gganimate

m0 <- lm(math ~ read + write + science + socst + race + female + ses, data = dta)
dta_m0 <- ggpredict(m0, terms = c("race", "female", "ses"))
plot(dta_m0) + labs(y = "Mean math score", x = "Race")

用原始資料作圖比較,有些只有一個點沒有error bar

ggplot(dta, aes(race, math, color = female))+
  stat_summary(fun.data = "mean_se", geom = "pointrange", na.rm = T)+ facet_grid(~ses)

EX3

dta3 <- read.table("kdt.csv", sep="",header=T)
head(dta3)

##   Test  Format Accuracy  SE
## 1  KDT Picture     93.7 0.9
## 2  KDT    Word     96.4 0.7
## 3  PPT Picture     90.6 1.0
## 4  PPT    Word     88.9 1.0

str(dta3)

## 'data.frame':    4 obs. of  4 variables:
##  $ Test    : Factor w/ 2 levels "KDT","PPT": 1 1 2 2
##  $ Format  : Factor w/ 2 levels "Picture","Word": 1 2 1 2
##  $ Accuracy: num  93.7 96.4 90.6 88.9
##  $ SE      : num  0.9 0.7 1 1

ggplot(dta3,aes(Test, Accuracy,fill=Format))+ 
  coord_cartesian(ylim=c(85,100))+
  geom_bar(stat="identity",position='dodge', colour="black")+
  geom_errorbar(aes(ymin = Accuracy-SE , ymax= Accuracy+SE),               width=.1,position=position_dodge(0.9))+
  scale_fill_manual(values=c( "gray", "gray30"))