Exercise 1_2_3: Statistical Software Files

In this set of exercises, we will work with files from statistical software. The first tasks are about importing data, while the later ones are about labelling and exporting.

1

Import the .sav version of the data from the German General Social Survey - ALLBUS 2021.

Clues

You need the haven package for this. The file should be stored in the data folder.

solution

library(haven)

allbus_2021 <- read_spss("./data/allbus_2021/ZA5280_v1-0-0.sav")

Your collaborators ask you to share the data. Unfortunately, they do not use R, SPSS or even Stata and, hence, asks you to export your data as a SAS file.

2

Export your data as a SAS file (its file extension is sas7bdat).

Clues

The haven package provides a function for writing such files that is called and works in a similar way as the corresponding function for importing data in this particular format.

solution

write_sas(allbus_2021, "Allbus_SAS_format.sas7bdat")

Unlike in flat files, such as CSV, the variables now have labels. If you aim to learn about them, you may want to have a look into the appendix for labelled data and then work on the following exercises. For this course, however, this is not necessary.

3 (optional)

Print the labels of the first ten variables in the data set.

Clues

You can use a function from the sjlabelled package for this. Remember that you can use [ ] ro subset columns/variables (we only want to print the labels for the first ten variables).

solution

library(sjlabelled)

get_label(allbus_2021[1:10])

##                                      za_nr                                        doi 
##                            "STUDIENNUMMER"                "DIGITAL OBJECT IDENTIFIER" 
##                                    version                                     respid 
##                                  "RELEASE"      "IDENTIFIKATIONSNUMMER DES BEFRAGTEN" 
##                                   substudy                                       mode 
##                               "TEILSTUDIE" "ERHEBUNGSMODUS DER ALLBUS-HAUPTBEFRAGUNG" 
##                                     splt21                                   eastwest 
##            "FRAGEBOGENSPLIT (A, B ODER C)" "ERHEBUNGSGEBIET (WOHNGEBIET): WEST - OST" 
##                                     german                                       ep01 
##           "DEUTSCHE STAATSANGEHOERIGKEIT?"     "WIRTSCHAFTSLAGE IN DEUTSCHLAND HEUTE"

Unfortunately, it’s all in German. Imagine you are a political scientist working on a publication in English, and you are interested in the variable pa01 (left-right self-placement). So you may want to consider translating the variable into English.

4 (optional)

Change the variable label of pa01 from “LINKS-RECHTS-SELBSTEINSTUFUNG, BEFR.” to “left-right self-placement”.

Clues

You can, again, use a function from sjlabelled for this.

solution

allbus_2021$pa01 <- 
  set_label(
    allbus_2021$pa01, 
    label = "left-right self-placement"
  )
get_label(allbus_2021$pa01)

## [1] "left-right self-placement"