36,543 rated or chosen profiles from 122 respondents, in up to 120 tasks of 3 profiles varying on 11 attributes.
| Description | Duke lab participants (eye-tracking, July 2019; n=122) each made 120 presidential vote choices in six blocks of 20 tasks, between 2 or 3 hypothetical candidates showing 5, 8 or 11 of 11 attributes (party and gender always shown). |
|---|---|
| Reference | Jenke, L., Bansak, K., Hainmueller, J., & Hangartner, D. (2021). Using eye-tracking to understand decision-making in conjoint experiments. Political Analysis, 29(1), 75-101. |
| DOI | 10.1017/pan.2020.11 |
| Licence | CC0 1.0 |
| Source data | https://doi.org/10.7910/DVN/TISLDL |
These notes describe the source data, which the IRW reproduces as released. All notes are listed on the Data Notes page.
| Respondents | 122 |
|---|---|
| Rows (one per respondent, task and profile) | 36,543 |
| Tasks per respondent (most) | 120 |
| Profiles per task | 3 |
| Attributes | 11 |
| 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 | yes (stated by the source): Attribute inclusion: party and gender are always shown; the other attributes in a 5- or 8-attribute block are a random subset (others '(not shown)'). No rules on level combinations stated. Level weights: race, military service and religion weighted toward more common groups (article 3.1; probabilities in Appendix A, not read); data: White 50%, Did not serve 67%, Catholic and Evangelical 25% each vs 17% for the other religions; other attributes near-uniform. |
| 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 | Your task is to decide which of the candidates you would vote for, for President, if you had to cast a vote. [choice, forced] |
|---|
attr_ageattr_genderattr_gun_controlattr_militaryattr_occupationattr_partyattr_political_experienceattr_raceattr_religionattr_same_sex_marriageattr_tax_wealthyattrpos_ageattrpos_genderattrpos_gun_controlattrpos_militaryattrpos_occupationattrpos_partyattrpos_political_experienceattrpos_raceattrpos_religionattrpos_same_sex_marriageattrpos_tax_wealthychoicecov_birth_yearcov_educationcov_gendercov_gun_positioncov_ideologycov_ideology_economiccov_ideology_socialcov_incomecov_know_robertscov_know_senate_termcov_know_speakercov_party_idcov_party_leancov_party_strengthcov_political_interestcov_political_understandingcov_racecov_ssmar_positioncov_student_statuscov_tax_positionidprofiletasktrial_blocktrial_block_itertrial_fixation_data
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("jenke_2021_eyetracking_candidates", source = "conj")
# Python
pip install irw
import irw
df = irw.fetch("jenke_2021_eyetracking_candidates", source="conj")
@article{Jenke_2020, title={Using Eye-Tracking to Understand Decision-Making in Conjoint Experiments}, volume={29}, ISSN={1476-4989}, url={http://dx.doi.org/10.1017/pan.2020.11}, DOI={10.1017/pan.2020.11}, number={1}, journal={Political Analysis}, publisher={Cambridge University Press (CUP)}, author={Jenke, Libby and Bansak, Kirk and Hainmueller, Jens and Hangartner, Dominik}, year={2020}, month=June, pages={75–101} }
| 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 |