13,803 rated or chosen profiles from 2,777 respondents, in up to 5 tasks of 1 profile varying on 8 attributes.
| Description | US adults (Ipsos KnowledgePanel, summer 2021; n=2,777) rated 5 single-profile political-conversation scenarios varying the partner's relationship, party, knowledge, news source, engagement, race and gender and the setting (face-to-face vs social media); the outcome is a 5-point rating of how likely they would be to express their true opinion. |
|---|---|
| Reference | Carlson, T. N., & Settle, J. E. (2023). Freedom of expression in interpersonal interactions. PS: Political Science & Politics, 56(2), 245-249. |
| DOI | 10.1017/S1049096522001342 |
| Licence | CC0 1.0 |
| Source data | https://doi.org/10.7910/DVN/FNHEVD |
These notes describe the source data, which the IRW reproduces as released. All notes are listed on the Data Notes page.
| Respondents | 2,777 |
|---|---|
| Rows (one per respondent, task and profile) | 13,803 |
| Tasks per respondent (most) | 5 |
| Profiles per task | 1 |
| Attributes | 8 |
| Topic | social identity interpersonal |
|---|---|
| A profile represents | scenario |
| Country | US |
| Language shown | en |
| Language of stored levels | en |
| Randomization restrictions | yes (stated by the source): conditional levels: the 'mainstream partisan' news level names Fox News for (Strong) Republican partners, MSNBC for (Strong) Democrat, Fox News or MSNBC at random (not recorded) for Independent; race Latino/a shown as Latino for male and Latina for female partners. Level weights: pre-registration says every attribute is shown with equal probability, but the data are clearly unequal: news partisan 66% / mainstream 17% / fringe 17% (expected 1/3 each); party Independent 24% vs ~19% for each other level |
| Task order | recorded in the deposit |
| Profile position | inferred from row order |
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. Task order or profile position was rebuilt from row order; analyses of position or task order should leave it out.
Outcomes:
| rating | Imagine that you were having a conversation about politics in the scenario described above. How likely would you be to express your true opinions in that conversation? [rating, 1 (Very unlikely) to 5 (Very likely)] |
|---|
attr_contextattr_engagementattr_genderattr_knowledgeattr_newsattr_partyattr_raceattr_relationshipcov_agecov_gendercov_local_conversationcov_party_idcov_political_participationcov_race_ethnicitycov_social_media_newscov_trust_mediaidprofileratingtask
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("carlson_2023_free_expression", source = "conj")
# Python
pip install irw
import irw
df = irw.fetch("carlson_2023_free_expression", source="conj")
@article{Carlson_2023, title={Freedom of Expression in Interpersonal Interactions}, volume={56}, ISSN={1537-5935}, url={http://dx.doi.org/10.1017/S1049096522001342}, DOI={10.1017/s1049096522001342}, number={2}, journal={PS: Political Science & Politics}, publisher={Cambridge University Press (CUP)}, author={Carlson, Taylor N. and Settle, Jaime E.}, year={2023}, month=Jan, pages={245–249} }
| 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 |