9,480 rated or chosen profiles from 1,189 respondents, in up to 4 tasks of 2 profiles varying on 6 attributes.
| Description | Caretakers whose child recently had diarrhoea (household survey, Karnataka and Bihar, India, n = 1,189) chose between two hypothetical healthcare providers in up to 4 choice sets; profiles always showed kindness, quality star rating and name type plus, depending on the version, the medicines given, the doctor fee or the travel time. Outcome is the chosen provider. |
|---|---|
| Reference | Wagner, Z., Mohanan, M., Mukherji, A., Zutshi, R., Patil, S., Krishnappa, J., Banerjee, S., & Sood, N. (2025). Investigating the know-do gap in antibiotics prescribing: Experimental evidence from India. Science Advances, 11. |
| DOI | 10.1126/sciadv.ady9868 |
| Licence | CC0 1.0 |
| Source data | https://doi.org/10.7910/DVN/5FXWC3 |
These notes describe the source data, which the IRW reproduces as released. All notes are listed on the Data Notes page.
| Respondents | 1,189 |
|---|---|
| Rows (one per respondent, task and profile) | 9,480 |
| Tasks per respondent (most) | 4 |
| Profiles per task | 2 |
| Attributes | 6 |
| Tasks with no profile chosen | 0 |
| Topic | health covid |
|---|---|
| A profile represents | product service |
| Country | IN |
| Language shown | unknown |
| Language of stored levels | en |
| Randomization restrictions | yes (stated by the source): fixed blocked design: 48 blocks x 4 choice sets, each block always the same profiles; three versions (randomized per household) show kindness, quality, name plus one of medicines prescribed, doctor fee or time to clinic; the other two are not shown |
| 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 | Choose one of the two providers (paraphrase; wording not deposited) [choice, forced] |
|---|
attr_kindnessattr_medicineattr_nameattr_priceattr_qualityattr_timechoiceidprofiletasktrial_blocktrial_choice_settrial_version
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 (no account needed with irw 1.4.0 or later) or Python (needs a free Redivis account; the first call opens a browser to sign in):
# R
install.packages("irw")
library(irw)
df <- irw_fetch("wagner_2025_provider_choice", source = "conj")
# Python
pip install irw
import irw
df = irw.fetch("wagner_2025_provider_choice", source="conj")
Or in Stata, which reads the same no-account CSV as the button above:
* Stata
net install irw, from("https://raw.githubusercontent.com/itemresponsewarehouse/Stata-pkg/main/")
irw_fetch wagner_2025_provider_choice, clear source(conj)
@article{Wagner_2025, title={Investigating the know-do gap in antibiotics prescribing: Experimental evidence from India}, volume={11}, ISSN={2375-2548}, url={http://dx.doi.org/10.1126/sciadv.ady9868}, DOI={10.1126/sciadv.ady9868}, number={37}, journal={Science Advances}, publisher={American Association for the Advancement of Science (AAAS)}, author={Wagner, Zachary and Mohanan, Manoj and Mukherji, Arnab and Zutshi, Rushil and Patil, Sumeet and Krishnappa, Jagadish and Banerjee, Somalee and Sood, Neeraj}, year={2025}, month=Sept }
| IRW version | v744 |
|---|---|
| Redivis dataset | irw_conjoint v5.0 |
| Redivis dataset DOI | 10.57761/x788-4838 |
| Manifest pin for this IRW version | v5.0 |
| Metadata source | irw_meta v78.0 |