1,521 rated or chosen profiles from 1,521 respondents, in up to 1 tasks of 1 profile varying on 2 attributes.

About this table

Description	1,521 attentive US partisans (YouGov panel, Jan 2024) read one vignette of a political rally at which attendees beat disruptors; the parties of attendees and disruptors (none named, own party beats opposing party, or the reverse) and the severity of the disruption (shouting or blocking the entrance) were randomized. Outcome: 1-7 justifiability of the beating.
Reference	Berntzen, L. E., Cappelen, C., Mason, L. H., & Midtbø, T. (2025). Mind the gap: Partisan bias in justifying political violence in the United States. Public Opinion Quarterly (conditionally accepted).
Licence	CC0 1.0
Source data	https://doi.org/10.7910/dvn/c2i1fk

Notes

Conjoint table (irw_conjoint_2). Factorial vignette, one per respondent (US, Jan 2024): which party's attendees beat which party's disruptors at a political rally (no party labels = '(not shown)'; own party beats opposing party; opposing party beats own party) x severity of the disruption (shouting vs blocking the venue entrance). rating = how justified the beating is, 1 = completely unjustified, 7 = completely justified. Kept only the authors' 1,521 attentive partisans (of 2,030); level text paraphrased from the codebook; the survey weight is kept although the authors analyse unweighted.

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

Design

Topic	crime justice policing
A profile represents	scenario
Country	US
Language shown	en
Language of stored levels	en
Task order	inferred from row order
Profile position	inferred from row order

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

Outcomes:

rating	How justified is it that the rally attendees beat the disruptors? (1-7: perceived justifiability of the beating; wording of the stem is not given, see codebook B1-B6) (paraphrase) [rating, 1 (Completely unjustified) to 7 (Completely justified)]

Columns

attr_disruption_severityattr_party_conditioncov_agecov_gender_codecov_income_codecov_partisan_strengthcov_party_idcov_race_codecov_region_codecov_survey_weightidprofileratingtask

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("berntzen_2025_political_violence", source = "conj")

# Python
pip install irw

import irw
df = irw.fetch("berntzen_2025_political_violence", source="conj")

How to cite

@misc{https://doi.org/10.7910/dvn/c2i1fk,
  doi = {10.7910/DVN/C2I1FK},
  url = {https://dataverse.harvard.edu/citation?persistentId=doi:10.7910/DVN/C2I1FK},
  author = {Berntzen, Lars Erik and Cappelen, Cornelius and Midtboe, Tor and Mason, Lilliana},
  keywords = {Social Sciences, Political violence, Survey experiment, Political behavior, United States},
  title = {Replication Data for: Mind the gap: Partisan bias in justifying poliical violence in the United States},
  publisher = {Harvard Dataverse},
  year = {2025}
}

Version and provenance

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

Respondents	1,521
Rows (one per respondent, task and profile)	1,521
Tasks per respondent (most)	1
Profiles per task	1
Attributes	2

berntzen_2025_political_violence