3,998 rated or chosen profiles from 1,999 respondents, in up to 1 tasks of 2 profiles varying on 9 attributes.

One of 4 tables from the same source; this one is the lucid2 table. The others are listed under Related tables.

About this table

Description	Democratic congressional primary candidate conjoint (Lucid study 2): 1,999 white respondents (Lucid, 2023; 1,852 Democrats) each chose between one Black and one White hypothetical candidate differing in race, age, job, political experience, endorsement and either policy positions or self-described ideology (between-subject versions in trial_version). Outcomes: vote choice plus six 0-1 trait ratings of each candidate.
Reference	Mikkelborg, A. C. (2025). White Democrats' growing support for Black politicians in the era of the "Great Awokening." American Political Science Review.
DOI	10.1017/S0003055425000097
Licence	CC0 1.0
Source data	https://doi.org/10.7910/DVN/LMBY8P

Notes

Conjoint table (irw_conjoint). choice: "Which candidate for Congress would you support in this Democratic Primary election?" (Candidate A/B, forced, no opt-out); one task per respondent. Restricted randomization: one candidate always Black, one White; no shared endorsement. Deposit contains white respondents only (all parties; the paper analyses Democrats: cov_party_id7 Lean, Mostly or Strong Democrat). Ratings stored 0-1 as deposited. Profile order inferred from row order. One respondent with no answers dropped.

These notes describe the source data, which the IRW reproduces as released. All notes are listed on the Data Notes page.

Design

Topic	candidate choice
A profile represents	person
Country	US
Language shown	en
Language of stored levels	en
Randomization restrictions	yes (stated by the source): race: one candidate always Black and the other White; the two candidates never share an endorsement (appendix also: nor the same priority if elected); other attributes equal probability
Task order	inferred from row order
Profile position	inferred from row order

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. Task order or profile position was rebuilt from row order; analyses of position or task order should leave it out.

Outcomes:

choice	Which candidate for Congress would you support in this Democratic Primary election? [choice, forced]
rating_demvotes	Please indicate how well you feel each phrase describes [CANDIDATE A/CANDIDATE B]: Would perform well with loyal Democratic voters [rating, 0 (Not well at all) to 1 (Extremely well)]
rating_goodchance	Please indicate how well you feel each phrase describes [CANDIDATE A/CANDIDATE B]: Has a good chance of winning the general election [rating, 0 (Not well at all) to 1 (Extremely well)]
rating_goodjob	Please indicate how well you feel each phrase describes [CANDIDATE A/CANDIDATE B]: Would do a good job if elected [rating, 0 (Not well at all) to 1 (Extremely well)]
rating_repsgroups	Please indicate how well you feel each phrase describes [CANDIDATE A/CANDIDATE B]: Represents historically underrepresented groups [rating, 0 (Not well at all) to 1 (Extremely well)]
rating_repsme	Please indicate how well you feel each phrase describes [CANDIDATE A/CANDIDATE B]: Represents people like me [rating, 0 (Not well at all) to 1 (Extremely well)]
rating_swingvotes	Please indicate how well you feel each phrase describes [CANDIDATE A/CANDIDATE B]: Would perform well with swing voters [rating, 0 (Not well at all) to 1 (Extremely well)]

Columns

attr_ageattr_climateattr_endorsementattr_experienceattr_healthcareattr_ideologyattr_jobattr_raceattr_reparationschoicecov_age01cov_educationcov_gendercov_hhinccov_ideocov_party_id7cov_regionidprofilerating_demvotesrating_goodchancerating_goodjobrating_repsgroupsrating_repsmerating_swingvotestasktrial_version

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.

Get the data

Download CSVno account needed Browse on Redivisexplore and query Croissant metadataHugging Face, Kaggle, OpenML

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("mikkelborg_2025_primary_lucid2", source = "conj")

# Python
pip install irw

import irw
df = irw.fetch("mikkelborg_2025_primary_lucid2", source="conj")

How to cite

@article{MIKKELBORG_2025, title={White Democrats’ Growing Support for Black Politicians in the Era of the “Great Awokening”}, volume={119}, ISSN={1537-5943}, url={http://dx.doi.org/10.1017/S0003055425000097}, DOI={10.1017/s0003055425000097}, number={4}, journal={American Political Science Review}, publisher={Cambridge University Press (CUP)}, author={MIKKELBORG, ANNA CAROLINE}, year={2025}, month=Feb, pages={1902–1920} }

Respondents	1,999
Rows (one per respondent, task and profile)	3,998
Tasks per respondent (most)	1
Profiles per task	2
Attributes	9
Tasks with no profile chosen	0

IRW version	v584
Redivis dataset	irw_conjoint v4.2
Redivis dataset DOI	10.57761/qkye-q530
Manifest pin for this IRW version	v4.2
Metadata source	irw_meta v44.0

mikkelborg_2025_primary_lucid2