5,020 rated or chosen profiles from 502 respondents, in up to 5 tasks of 2 profiles varying on 7 attributes.

About this table

Description	502 Romanian eligible voters (non-probability sample in 30 localities, 2016, before the local elections) each chose between 5 pairs of hypothetical mayoral candidates described by 7 attributes: gender, incumbency, corruption investigation, vote buying, threats to non-supporters, policy pledges and income. Outcome: vote choice, with a neither option.
Reference	Mares, I., & Visconti, G. (2020). Voting for the lesser evil: Evidence from a conjoint experiment in Romania. Political Science Research and Methods, 8(2), 315-328.
DOI	10.1017/psrm.2019.12
Licence	CC0 1.0
Source data	https://doi.org/10.7910/DVN/MJKCHZ

Notes

Conjoint table (irw_conjoint). choice = which candidate the respondent would vote for (authors' 'Electoral Choice'; question wording not deposited), with a third, undocumented response coded 3 in 257 of 2,510 tasks treated as an opt-out (both profiles 0). Restriction: every pair shows one incumbent mayor and one candidate without mayoral experience. Attribute text is the appendix's English version. N = 502 as in the paper.

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

Size and shape

Respondents	502
Rows (one per respondent, task and profile)	5,020
Tasks per respondent (most)	5
Profiles per task	2
Attributes	7
Tasks with no profile chosen	257

Design

Topic	candidate choice
A profile represents	person
Country	RO
Language shown	unknown
Language of stored levels	en
Randomization restrictions	yes (stated by the source): every pair has exactly one incumbent ('currently serving as mayor') and one candidate without experience as mayor; order within the pair randomized (appendix Table A1 fn 1); no rule stated for other attributes
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	Choice between the two candidates in a local (mayoral) election, 'Electoral Choice' (paraphrase) [choice, opt-out allowed]

Columns

attr_experienceattr_genderattr_incomeattr_investigationattr_public_policyattr_threatattr_vote_buyingchoicecov_agecov_birth_yearcov_countycov_educationcov_ethnicity_codecov_gendercov_income_codecov_party_codecov_urbanidprofiletask

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("mares_2020_lesser_evil", source = "conj")

# Python
pip install irw

import irw
df = irw.fetch("mares_2020_lesser_evil", source="conj")

How to cite

@article{Mares_2019, title={Voting for the lesser evil: evidence from a conjoint experiment in Romania}, volume={8}, ISSN={2049-8489}, url={http://dx.doi.org/10.1017/psrm.2019.12}, DOI={10.1017/psrm.2019.12}, number={2}, journal={Political Science Research and Methods}, publisher={Cambridge University Press (CUP)}, author={Mares, Isabela and Visconti, Giancarlo}, year={2019}, month=June, pages={315–328} }

Version and provenance

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