9,486 rated or chosen profiles from 3,162 respondents, in up to 3 tasks of 1 profile varying on 12 attributes.
| Description | 3,162 US adults (YouGov, December 2024) each read three newspaper-style vignettes about a state-legislature candidate, randomizing the candidate's photo/name (race, gender, age), party, and (in ~80% of vignettes) a controversial remark (type, target, wording), its timing, planning and history, and the politician's response. Outcomes: would ever consider voting (yes/no), 0-100 thermometer, 1-7 objectionableness. |
|---|---|
| Reference | Gift, T., & Lastra-Anadón, C. X. (2025). Can politicians say that? What shapes public responses to speech scandals? Journal of Experimental Political Science. |
| DOI | 10.1017/XPS.2025.10010 |
| Licence | CC0 1.0 |
| Source data | https://doi.org/10.7910/DVN/DUVWXQ |
These notes describe the source data, which the IRW reproduces as released. All notes are listed on the Data Notes page.
| Respondents | 3,162 |
|---|---|
| Rows (one per respondent, task and profile) | 9,486 |
| Tasks per respondent (most) | 3 |
| Profiles per task | 1 |
| Attributes | 12 |
| Tasks with no profile chosen | 4,786 |
| Topic | candidate choice |
|---|---|
| A profile represents | person |
| Country | US |
| Language shown | en |
| Language of stored levels | en |
| Randomization restrictions | yes (stated by the source): Remark, target, quote version, response, timing, planning and history appear only in treated vignettes (control = clean description, ~20%); YouGov avoided showing the same photo or quote twice to a respondent (article fn. 4, 8). Control share ~20% (split_1_ext), so the remark attribute's levels are not equally likely. |
| 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 | If you had to make a choice without knowing more, would you ever consider voting for this candidate? [choice, opt-out allowed] |
|---|---|
| rating_objectionable | On a scale of 1 to 7, how objectionable do you think the candidate's behavior is in this case? [rating, 1 (Not at all objectionable) to 7 (The most objectionable I can imagine)] |
| rating_thermometer | Feeling thermometer toward the candidate, 0-100 from coldness to warmth (paraphrase) [rating, 0 (coldness) to 100 (warmth)] |
attr_ageattr_genderattr_historyattr_nameattr_partyattr_plannedattr_quote_versionattr_raceattr_remarkattr_responseattr_targetattr_timingchoicecov_birth_yearcov_educ_codecov_gendercov_ideo5_codecov_omnibuscov_pid3_codecov_pid7_codecov_race_codecov_survey_weightidprofilerating_objectionablerating_thermometertask
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("gift_2025_speech_scandals", source = "conj")
# Python
pip install irw
import irw
df = irw.fetch("gift_2025_speech_scandals", source="conj")
@article{Gift_2025, title={Can Politicians Say That? What Shapes Public Responses to Speech Scandals}, volume={13}, ISSN={2052-2649}, url={http://dx.doi.org/10.1017/XPS.2025.10010}, DOI={10.1017/xps.2025.10010}, number={1}, journal={Journal of Experimental Political Science}, publisher={Cambridge University Press (CUP)}, author={Gift, Thomas and Lastra-Anadón, Carlos X.}, year={2025}, month=Aug, pages={91–107} }
| 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 |