We create a minimal question bank and a codebook for the most important variables of the GESIS Eurobarometer microdata files. These are the variables that are likely to be needed in every harmonization job.
data("eurobarometer_basic_codebook")
print(eurobarometer_basic_codebook)
#> Title ZACAT var_name_orig var_label_orig
#> 1 Eurobarometer 67.1 (Feb-Mar 2007) ZA4529 v5 ID SERIAL NUMBER
#> var_name_target var_label_target zacat_doi
#> 1 uniqid GESIS UNIQUE ID [ID SERIAL NUMBER] 10.4232/1.10983Identification
We place the survey-level metadata, such as the title of the survey, the DOI of the survey file, into a dataset metadata.
The most important variable that needs to be found is the unique case
id for every response. In more recent Eurobarometers, this is always
called uniqid. In older files, it needs to be looked up.
The installed version of eurobarometer will contain this
crucial information.
Protocol variables
Our basic codebook contains one metadata variable, the date of the
interview. In recent Eurobarometer files, this is the p1
variable. In earlier files, it maybe necessary to look up this file.
Weight variables
In recent Eurobarometer files, most, but not all weight variables are
consistently named. We locate the two important variables,
w1 (basic post-stratification weight) and wex
(weight extrapolated on population).
Basic demography variables
We will include in the basic questionbank, codetable, and we will separately release in processed forms the most important socio-demographic variables.
get_demography_schema()
#> # A tibble: 7 × 3
#> var_name_orig var_name_target class_target
#> <chr> <chr> <chr>
#> 1 uri uri character
#> 2 d11 age_exact numeric
#> 3 d25 eb_type_community declared
#> 4 d7 marital_status declared
#> 5 d8 age_education declared
#> 6 d15a occupation declared
#> 7 d15b occupation_last declaredA tutorial will follow on how to use them.