This data frame contains respondent-level data from a number of recent data collections.
The public release version of these data has been trimmed in order to preserve respondent
privacy. Covariates retained, where available, are gender identification, racial/ethnic
identification, and age. Full versions of these data may be available from the authors.
Call dict_info() for citation information.
Format
A data frame with 794449 rows and 16 variables:
- dataset
the data collection effort the observation belongs to
- context
context in which dataset was collect (usually country)
- year
year dataset collected/published
- userid
individual identifier
- gender
respondent's gender identity. Male/female except in occupations data sets, where genderqueer/nonbinary and other are options as well.
- age
respondent's age or age bracket
- raceeth
respondent's racial/ethnic identification (relevant to artifactmods2022, humanvalues2022, products2022, occs2019, occs2020, occs2021)
- race1
respondent's primary racial identification (relevant to dukecommunity2015, dukestudent2015, uga2015, usmturk2015, usstudent2015, usfullsurveyor2015)
- race2
respondent's secondary racial identification (relevant to dukecommunity2015, dukestudent2015, uga2015, usmturk2015, usstudent2015, usfullsurveyor2015)
- hisp
whether a respondent identifies as Hispanic (relevant to uga2015, mturk2015, usstudent2015, usfullsurveyor2015)
- term
dictionary term
- component
the term's type
- instcodes
institution code
- E
evaluation rating
- P
potency rating
- A
activity rating
...
Details
Please note that covariates may not be comparable between data sets. Different data collections provided different response options for these questions. Only very minimal recoding has been done for the version included here. Racial and ethnic identification variables show the most differences between data sets. If you wish to use these variables for research, you should contact the data set authors in order to obtain questionnaire information.