1,534 rated or chosen profiles from 1,534 respondents, in up to 1 tasks of 1 profile varying on 2 attributes.
| Description | Household heads along the Jamuna river in rural Bangladesh (panel wave 1, May-June 2021, face-to-face) heard one of six hypothetical riverbank-erosion and flood scenarios (3 damage levels x 2 future-risk levels) and said whether they would stay in their village or move. 1,534 respondents; one stay/move choice each. |
|---|---|
| Reference | Rudolph, L. (2026). Vignette experiments can replicate actual behavioral intent and partly actual behavior: Panel evidence on environmental migration from Bangladesh. Journal of Experimental Political Science, 1-22. |
| DOI | 10.1017/XPS.2026.10035 |
| Licence | CC0 1.0 |
| Source data | https://doi.org/10.7910/DVN/CON75O |
These notes describe the source data, which the IRW reproduces as released. All notes are listed on the Data Notes page.
| Respondents | 1,534 |
|---|---|
| Rows (one per respondent, task and profile) | 1,534 |
| Tasks per respondent (most) | 1 |
| Profiles per task | 1 |
| Attributes | 2 |
| Tasks with no profile chosen | 1,205 |
| Topic | immigration asylum, climate energy environment |
|---|---|
| A profile represents | scenario |
| Country | BD |
| Language shown | bn |
| Language of stored levels | en |
| Randomization restrictions | yes (stated by the source): Six fixed vignettes (3 damage x 2 risk), one per respondent, each 16.6% of the sample. The situation and outlook texts are not independently randomized: the no-damage situation text differs by risk (normal year vs household spared), and the high-risk outlook in the no-damage cell drops the word 'any'. Use trial_design_cell for the damage x risk factors. |
| 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 | If you were in this situation, would you stay here in your present village or would you prefer to move to another place? (translated) [choice, opt-out allowed] |
|---|
attr_outlookattr_situationchoicecov_agecov_attachmentcov_districtcov_educationcov_gendercov_land_sizecov_risk_prefcov_zoneidprofiletasktrial_dependstrial_design_cell
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("rudolph_2026_migration_vignette", source = "conj")
# Python
pip install irw
import irw
df = irw.fetch("rudolph_2026_migration_vignette", 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 rudolph_2026_migration_vignette, clear source(conj)
@article{Rudolph_2026, title={Vignette Experiments Can Replicate Actual Behavioral Intent and Partly Actual Behavior: Panel Evidence on Environmental Migration from Bangladesh}, ISSN={2052-2649}, url={http://dx.doi.org/10.1017/XPS.2026.10035}, DOI={10.1017/xps.2026.10035}, journal={Journal of Experimental Political Science}, publisher={Cambridge University Press (CUP)}, author={Rudolph, Lukas}, year={2026}, month=June, pages={1–22} }
| 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 |