48,088 rated or chosen profiles from 4,812 respondents, in up to 5 tasks of 2 profiles varying on 7 attributes.

About this table

Description	4,812 self-identified Democrats and Republicans (US quota-sampled panel) chose between two randomly generated communities in each of 5 tasks, described by housing cost, partisan composition, racial composition, school quality, commute, crime rate and type of place. Outcome: preferred community of the pair (forced choice).
Reference	Mummolo, J., & Nall, C. (2017). Why partisans do not sort: The constraints on political segregation. The Journal of Politics, 79(1), 45-59.
DOI	10.1086/687569
Licence	CC0 1.0
Source data	https://doi.org/10.7910/dvn/edgrdc

Notes

Conjoint table (irw_conjoint_2). Forced choice between two randomly generated communities (Community A/B), 5 tasks per respondent, 7 attributes fully randomized; choice = preferred community (paraphrase of the paper), no opt-out. 16 task-pairs with missing choice dropped. Housing cost levels are 15/30/40 percent in the data, 15/25/40 in the paper's text; the deposit's second experiment (single trait labels, no profiles) is not included.

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	4,812
Rows (one per respondent, task and profile)	48,088
Tasks per respondent (most)	5
Profiles per task	2
Attributes	7
Tasks with no profile chosen	0

Design

Topic	housing neighbourhoods
A profile represents	place
Country	US
Language shown	en
Language of stored levels	en
Task order	recorded in the deposit
Profile position	recorded in the deposit

Outcomes:

choice	Which community would the respondent choose, Community A or Community B (paraphrase: choose the preferred community from each pair) [choice, forced]

Columns

attr_commuteattr_crimeattr_housing_costattr_partisan_voteattr_racial_compositionattr_school_qualityattr_type_of_placechoicecov_agecov_education_yearscov_gendercov_income_1000cov_party_leancov_race_blackcov_race_whiteidprofiletask

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("mummolo_2017_community_choice", source = "conj")

# Python
pip install irw

import irw
df = irw.fetch("mummolo_2017_community_choice", source="conj")

How to cite

@article{Mummolo_2017, title={Why Partisans Do Not Sort: The Constraints on Political Segregation}, volume={79}, ISSN={1468-2508}, url={http://dx.doi.org/10.1086/687569}, DOI={10.1086/687569}, number={1}, journal={The Journal of Politics}, publisher={University of Chicago Press}, author={Mummolo, Jonathan and Nall, Clayton}, year={2017}, month=Jan, pages={45–59} }

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