16  常用技巧

16.1 类型转换

library(tidyverse)
data<-data.frame(id=1:12,age=c(0.5,1.2,3,4,5,6),sex=c('男','女'),virus=c('1','0'))
data

data_new<-data %>%
  mutate(id=as.character(id), #转为字符型
         virus=as.numeric(virus)) #转为数值型
data_new

16.2 连续转分类/水平转换

提示

ifelse()和case_when()函数二选一。简单二分类用ifelse()即可，复杂情况推荐用case_when()可避免多个ifelse()嵌套结构。

data_new %>%
  mutate(agegroup=ifelse(age<1,'0~1y',
                         ifelse(age<2,'1~2y','2y+'))) %>%
  mutate(agegroup2=case_when(age<1~'0~1y',
                             age<2~'1~2y',
                             T~'2y+')) %>%
  mutate(sex2=ifelse(sex=='男','Male','Female'),
         sex3=case_when(sex=='男'~'Male',
                        T~'Female'))

提示

男性和女性贫血判定的界值不同，根据性别和界值判断是否贫血。

data<-data.frame(
  sex=rep(c('Female','Male'),3),
  Hb=runif(6,100,150) %>% round())

data

data %>%
  mutate(cutoff=ifelse(sex=='Female',115,130),
         Hb_2=ifelse(Hb<=cutoff,'abnormal','normal')) %>%
  select(-cutoff)

data %>%
  mutate(Hb_2=case_when(
    Hb<=115 & sex=='Female'~'abnormal',
    Hb<=130 & sex=='Male'~'abnormal',
    T~'normal'
  ))

16.3 按层划分连续转分类

提示

分别针对男性和女性按照血压/血脂/血糖高低划分为四组(Q1,Q2,Q3,Q4)

library(tidyverse)
df <- data.frame(
  gender = rep(c("Male", "Female"), each = 200),
  blood_pre = round(runif(400, min = 90, max = 140)),
  blood_lip = round(runif(400, min = 1.5, max = 5.0),1),
  blood_sug = round(runif(400, min = 3.5, max = 7.5),1)
)

# 定义函数计算四分位数
gen_quantile <- function(x) {
  quantiles <- quantile(x, probs = c(0, 0.25, 0.5, 0.75, 1), na.rm = TRUE)
  cut(x, breaks = quantiles, include.lowest = TRUE, labels = c("Q1", "Q2", "Q3", "Q4"))
}

# 按性别分组并计算四分位数
df %>%
  group_by(gender) %>%
  mutate(pre_group = gen_quantile(blood_pre),
         lip_group = gen_quantile(blood_lip),
         sug_group = gen_quantile(blood_sug))