34,832 rated or chosen profiles from 2,177 respondents, in up to 8 tasks of 2 profiles varying on 7 attributes.
| Description | Paired conjoint: 2,177 US MTurk respondents (October 2021) compared 8 pairs of hypothetical welfare programs (EITC- or TANF-style descriptions, negative information about inadequacy and fraud, funding level of government, cost, participants, time limit), choosing one and rating support for each. |
|---|---|
| Reference | Qi, H., & Haselswerdt, J. (2024). Ideology, information, and social welfare preferences. American Politics Research, 52(1), 41-51. |
| DOI | 10.1177/1532673X231206151 |
| Licence | CC0 1.0 |
| Source data | https://doi.org/10.7910/DVN/NE1WLH |
These notes describe the source data, which the IRW reproduces as released. All notes are listed on the Data Notes page.
| Respondents | 2,177 |
|---|---|
| Rows (one per respondent, task and profile) | 34,832 |
| Tasks per respondent (most) | 8 |
| Profiles per task | 2 |
| Attributes | 7 |
| Tasks with no profile chosen | 0 |
| Topic | tax welfare economy |
|---|---|
| A profile represents | policy |
| Country | US |
| Language shown | en |
| Language of stored levels | en |
| Randomization restrictions | yes (stated by the source): EITC descriptions appear only with EITC problem statements or 'None', TANF descriptions only with TANF ones or 'None' (authors' constraint(cat_a_desc#cat_a_prob1#cat_a_prob2); 8 empty cells each); within problem 1 and problem 2 'None' has share 0.34 and each family-specific level ~0.17, consistent with equal draws from the three levels allowed for the profile's program family; other attributes near-equal |
| Task order | inferred from row order |
| 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:
| choice | unknown (forced choice between the two programs; source program_choice 'Chose this program' / 'Chose other program'; wording not in the deposit) [choice, forced] |
|---|---|
| rating | unknown (program support for each program; source program_support, 'Program support'; wording not in the deposit) [rating, 1 (unknown) to 7 (unknown)] |
attr_costattr_descriptionattr_fundingattr_participantsattr_problem_1attr_problem_2attr_time_limitchoicecov_agecov_attention_pass_1cov_attention_pass_2cov_attention_pass_3cov_attention_pass_4cov_duration_seccov_educationcov_employcov_faminccov_gendercov_ideologycov_news_attentioncov_party_idcov_party_id7cov_raceidprofileratingtask
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("qi_2024_welfare_programs", source = "conj")
# Python
pip install irw
import irw
df = irw.fetch("qi_2024_welfare_programs", source="conj")
@article{Qi_2023, title={Ideology, Information, and Social Welfare Preferences}, volume={52}, ISSN={1552-3373}, url={http://dx.doi.org/10.1177/1532673X231206151}, DOI={10.1177/1532673x231206151}, number={1}, journal={American Politics Research}, publisher={SAGE Publications}, author={Qi, Hang and Haselswerdt, Jake}, year={2023}, month=Oct, pages={41–51} }
| IRW version | v607 |
|---|---|
| Redivis dataset | irw_conjoint_2 v1.5 |
| Redivis dataset DOI | 10.57761/g2hc-zw49 |
| Manifest pin for this IRW version | v1.4 |
| Metadata source | irw_meta v47.0 |