foreign

R 数据导入进阶

Filip Schouwenaars

Instructor, DataCamp

foreign

R 核心团队
一致性较弱
覆盖面很广
各类外部数据格式
SAS、STATA、SPSS、Systat、Weka …

install.packages("foreign")
library(foreign)

SAS

不能导入 .sas7bdat
仅支持 SAS 库：.xport
sas7bdat 包

STATA

STATA 5 到 12
read.dta() - read.dta()

read.dta(file, 
         convert.factors = TRUE,
         convert.dates = TRUE,
         missing.type = FALSE)

read.dta()

ontime <- read.dta("ontime.dta")
ontime

         Airline March_1999 June_1999 August_1999
1            TWA       84.4      69.4        85.0
2      Southwest       80.3      77.0        80.4
3      Northwest       80.8      75.1        81.0
4       American       72.7      65.1        78.3
5          Delta       78.7      72.2        77.7
6    Continental       79.3      68.4        75.1
7         United       78.6      69.2        71.6
8     US Airways       73.6      68.9        70.1
9         Alaska       71.9      75.4        64.4
10 American West       76.5      70.3        62.5

read.dta()

ontime <- read.dta("ontime.dta")
str(ontime)

convert.factors 默认 TRUE

'data.frame':    10 obs. of  4 variables:
 $ Airline    : Factor w/ 10 levels "Alaska",..: 8 7 6 2 5 4 ...
 $ March_1999 : num  84.4 80.3 80.8 72.7 78.7 79.3 78.6 ...
 $ June_1999  : num  69.4 77 75.1 65.1 72.2 68.4 69.2 68.9 ...
 $ August_1999: num  85 80.4 81 78.3 77.7 75.1 71.6 70.1 ...
 - attr(*, "datalabel")= chr "Written by R.              "
 - attr(*, "time.stamp")= chr ""
 - attr(*, "formats")= chr  "%9.0g" "%9.0g" "%9.0g" "%9.0g"
 - attr(*, "types")= int  108 100 100 100
 - attr(*, "val.labels")= chr  "Airline" "" "" ""
 - attr(*, "var.labels")= chr  "Airline" "March_1999" ...
 - attr(*, "version")= int 7
 - attr(*, "label.table")=List of 1
  ..$ Airline: Named int  1 2 3 4 5 6 7 8 9 10
  .. ..- attr(*, "names")= chr  "Alaska" "American" ...

read.dta() - convert.factors

ontime <- read.dta("ontime.dta", convert.factors = FALSE)
str(ontime)

'data.frame': 10 obs. of  4 variables:
 $ Airline    : int  8 7 6 2 5 4 9 10 1 3
 $ March_1999 : num  84.4 80.3 80.8 72.7 78.7 79.3 78.6 ...
 $ June_1999  : num  69.4 77 75.1 65.1 72.2 68.4 69.2 68.9 ...
 $ August_1999: num  85 80.4 81 78.3 77.7 75.1 71.6 70.1 ...
 - attr(*, "datalabel")= chr "Written by R.              "
 - attr(*, "time.stamp")= chr ""
 - attr(*, "formats")= chr  "%9.0g" "%9.0g" "%9.0g" "%9.0g"
 - attr(*, "types")= int  108 100 100 100
 - attr(*, "val.labels")= chr  "Airline" "" "" ""
 - attr(*, "var.labels")= chr  "Airline" "March_1999" ...
 - attr(*, "version")= int 7
 - attr(*, "label.table")=List of 1
  ..$ Airline: Named int  1 2 3 4 5 6 7 8 9 10
  .. ..- attr(*, "names")= chr  "Alaska" "American" ...

read.dta() - 其他参数

read.dta(file, 
         convert.factors = TRUE,
         convert.dates = TRUE,
         missing.type = FALSE)

convert.factors：将带标签的 STATA 值转为 R 因子

convert.dates：将 STATA 日期/时间转为 Date 和 POSIXct

missing.type：

若为 FALSE，将所有缺失类型转为 NA
若为 TRUE，在属性中保存缺失类型

SPSS

read.spss()

read.spss(file,
          use.value.labels = TRUE,
          to.data.frame = FALSE)

use.value.labels：将带标签的 SPSS 值转为 R 因子

to.data.frame：返回数据框而非列表

trim.factor.names

trim_values

use.missings

¡Vamos a practicar!

R 数据导入进阶