27,402 rated or chosen profiles from 1,971 respondents, in up to 7 tasks of 2 profiles varying on 5 attributes.
| Description | Mayoral-candidate conjoint on a nationally representative US sample fielded through Lucid Theorem on 28 July 2021: 1,971 respondents x 7 paired tasks x 2 profiles (27,402 rows), forced choice on "Suppose you lived in that city. Which of these two candidates would you be likely to vote for mayor?". Five attributes, twelve levels: gender (Man / Woman), leadership style (Decisive / Looks for compromises), policy priority (Build new schools / Increase recycling / Keep taxes low), previous term (Kept all of their promises / Broke most of their promises) and occupation (Businessperson / Lawyer / Teacher). Attribute order was randomized once per respondent and held across that respondent's seven tasks; attrpos_* carry it. Covariates: respondent gender, age, education, race, party identification and strength, which of the three policy issues matters most and next, the attribute the respondent says mattered most, survey duration, and the two attention checks as cov_attention_pass_1 and _2 -- the paper's analysis sample of 1,495 is this table filtered to respondents who pass both and finished. Conjoint layout (data/conjoint/README.md). |
|---|---|
| Reference | Sevi, S., Crabtree, C., & Blais, A. Do voters punish women politicians more? [manuscript; the deposit names American Politics Research]. Replication data: OSF, https://osf.io/mgkup/ |
| Licence | CC BY 4.0 |
| Source data | https://osf.io/mgkup/ |
These notes describe the source data, which the IRW reproduces as released. All notes are listed on the Data Notes page.
| Respondents | 1,971 |
|---|---|
| Rows (one per respondent, task and profile) | 27,402 |
| Tasks per respondent (most) | 7 |
| Profiles per task | 2 |
| Attributes | 5 |
| Tasks with no profile chosen | 0 |
| Topic | candidate choice |
|---|---|
| A profile represents | person |
| Country | US |
| Language shown | en |
| Language of stored levels | en |
| Randomization restrictions | unknown |
| Task order | recorded in the deposit |
| Profile position | recorded in the deposit |
Outcomes:
| choice | Suppose you lived in that city. Which of these two candidates would you be likely to vote for mayor? [choice, forced] |
|---|
attr_genderattr_leadership_styleattr_occupationattr_policy_priorityattr_previous_termattrpos_genderattrpos_leadership_styleattrpos_occupationattrpos_policy_priorityattrpos_previous_termchoicecov_agecov_attention_pass_1cov_attention_pass_2cov_considerationcov_duration_seccov_educationcov_finishedcov_gendercov_issue_firstcov_issue_secondcov_party_idcov_party_strengthcov_raceidprofiletask
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 (no account needed with irw 1.4.0 or later) or Python (needs a free Redivis account; the first call opens a browser to sign in):
# R
install.packages("irw")
library(irw)
df <- irw_fetch("sevi_2025_promise_keeping", source = "conj")
# Python
pip install irw
import irw
df = irw.fetch("sevi_2025_promise_keeping", source="conj")
Or in Stata, which reads the same no-account CSV as the button above:
* Stata
net install irw, from("https://raw.githubusercontent.com/itemresponsewarehouse/Stata-pkg/main/")
irw_fetch sevi_2025_promise_keeping, clear source(conj)
@dataset{sevi_crabtree_blais_voters,
author = {Sevi, S. and Crabtree, C. and Blais, A.},
title = {Do voters punish women politicians more? {R}eplication data},
year = {n.d.},
publisher = {Open Science Framework},
url = {https://osf.io/mgkup/},
doi = {10.17605/OSF.IO/MGKUP}
}
| IRW version | v744 |
|---|---|
| Redivis dataset | irw_conjoint_2 v4.0 |
| Redivis dataset DOI | 10.57761/npz5-mq61 |
| Manifest pin for this IRW version | v4.0 |
| Metadata source | irw_meta v78.0 |