145,296 rated or chosen profiles from 6,054 respondents, in up to 12 tasks of 2 profiles varying on 7 attributes.
| Description | Online samples in the United States, South Korea and Poland (n = 6,054) chose between two hypothetical candidates in each of 12 tasks; candidates varied in age, gender, background, party, views on tax and spending, gender policy and anti- or pro-democratic positions (polling stations, court rulings, prosecuting journalists). |
|---|---|
| Reference | Claassen, C. (2026). Voting for democracy? Testing political-cultural constraints on democratic backsliding. Public Opinion Quarterly. |
| DOI | 10.1093/poq/nfag066 |
| Licence | CC0 1.0 |
| Source data | https://doi.org/10.7910/DVN/N9BFTG |
These notes describe the source data, which the IRW reproduces as released. All notes are listed on the Data Notes page.
| Respondents | 6,054 |
|---|---|
| Rows (one per respondent, task and profile) | 145,296 |
| Tasks per respondent (most) | 12 |
| Profiles per task | 2 |
| Attributes | 7 |
| Tasks with no profile chosen | 0 |
| Topic | candidate choice, democracy institutions |
|---|---|
| A profile represents | person |
| Country | US, KR, PL |
| Language shown | en, ko, pl |
| Language of stored levels | en |
| Randomization restrictions | yes (stated by the source): the political-system attribute is drawn so that one candidate of a pair has an anti-democratic level (cut polling stations; ignore court rulings; prosecute journalists) and the other a pro-democratic level (verified: exactly one odd and one even code in all 72,648 tasks); gender Male 0.8 / Female 0.2; party US 0.5/0.5, KR 0.5/0.5, PL PiS 0.35, KO 0.35, Kon 0.15, Lewica 0.15; other attributes uniform |
| Task order | recorded in the deposit |
| Profile position | recorded in the deposit |
Because the attribute levels were not randomized independently, an estimator has to account for the restriction; a plain difference in means across levels can mislead.
Outcomes:
| choice | Which candidate do you prefer? (Candidate 1 / Candidate 2) [choice, forced] |
|---|
attr_ageattr_backgroundattr_genderattr_gender_policyattr_partyattr_system_changeattr_tax_spendingchoicecov_agecov_countrycov_duration_seccov_educationcov_genderidprofiletask
The layout (one row per respondent, task and profile; choice/rating outcomes; attr_ columns holding each attribute level as displayed) is defined in the IRW conjoint standard. What the outcomes asked is under Design above.
Or load it directly in R or Python (both need a free Redivis account; the first call opens a browser to sign in):
# R
install.packages("remotes")
remotes::install_github("itemresponsewarehouse/Rpkg")
library(irw)
df <- irw_fetch("claassen_2026_democratic_backsliding", source = "conj")
# Python
pip install irw
import irw
df = irw.fetch("claassen_2026_democratic_backsliding", source="conj")
@article{Claassen_2026, title={Voting for Democracy? Testing Political-Cultural Constraints on Democratic Backsliding}, ISSN={1537-5331}, url={http://dx.doi.org/10.1093/poq/nfag066}, DOI={10.1093/poq/nfag066}, journal={Public Opinion Quarterly}, publisher={Oxford University Press (OUP)}, author={Claassen, Christopher}, year={2026}, month=Sept }
| IRW version | v584 |
|---|---|
| Redivis dataset | irw_conjoint_2 v1.4 |
| Redivis dataset DOI | 10.57761/hk97-wv30 |
| Manifest pin for this IRW version | v1.4 |
| Metadata source | irw_meta v44.0 |