14,878 rated or chosen profiles from 1,497 respondents, in up to 5 tasks of 2 profiles varying on 7 attributes.
| Description | Armenian adults (face-to-face tablet survey, 2023, n = 1,497) compared 5 pairs of fictional corruption scenarios varying type of corruption, sector, official's rank and gender, amount, use of funds and sharing, choosing which harms their family, the economy, political trust most, which is more morally wrong and which is more frequent. |
|---|---|
| Reference | Erlich, A., Gans-Morse, J., Nichter, S., & Holverscheid, A. (2025). What corruption is most harmful? Unbundling citizen perceptions. World Development, 194, 107001. |
| DOI | 10.1016/j.worlddev.2025.107001 |
| Licence | CC0 1.0 |
| Source data | https://doi.org/10.7910/DVN/3HHOMK |
These notes describe the source data, which the IRW reproduces as released. All notes are listed on the Data Notes page.
| Respondents | 1,497 |
|---|---|
| Rows (one per respondent, task and profile) | 14,878 |
| Tasks per respondent (most) | 5 |
| Profiles per task | 2 |
| Attributes | 7 |
| Topic | corruption accountability |
|---|---|
| A profile represents | scenario |
| Country | AM |
| Language shown | hy |
| Language of stored levels | hy |
| Randomization restrictions | yes (stated by the source): 'gives corrupt favors to family members' and 'uses public resources to benefit own firm' only with personal use and 'keeps all corrupt funds'; political use never with 'keeps all corrupt funds' (design1.csv; authors estimate within constrained combinations, 05_constrained_lm.R). Resulting shares: these two types 6% each vs 29-30% for the others; personal use 64%; keeps-all 29% |
| 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_economy | According to you, which of the scenarios will probably harm Armenia's economy more? Scenario A or Scenario B. (translated) [choice, forced] |
|---|---|
| choice_frequent | And in the end, please tell me, which of these two scenarios do you think happens most often in Armenia? Scenario A or Scenario B. (translated) [choice, forced] |
| choice_moral | And now, please tell me in which of the following scenarios do you think the official's action is more morally wrong? Scenario A or Scenario B. (translated) [choice, forced] |
| choice_personal | First, please indicate which of the scenarios is likely to harm you and your family the mostÕ¯ Scenario A or Scenario B. (translated) [choice, forced] |
| choice_trust | Would you indicate which of the scenarios would most likely reduce your confidence in the political system of Armenia? Scenario A or Scenario B. (translated) [choice, forced] |
attr_act_aloneattr_official_genderattr_official_rankattr_official_useattr_sectorattr_size_corruptattr_typechoice_economychoice_frequentchoice_moralchoice_personalchoice_trustcov_agecov_childrencov_corruption_frequencycov_educationcov_family_financescov_gendercov_harm_economycov_harm_personalcov_harm_trustcov_incomecov_maritalcov_regioncov_residencecov_survey_weightidprofiletask
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("erlich_2025_corruption_harms", source = "conj")
# Python
pip install irw
import irw
df = irw.fetch("erlich_2025_corruption_harms", source="conj")
@article{Erlich_2025, title={What corruption is most harmful? Unbundling citizen perceptions}, volume={194}, ISSN={0305-750X}, url={http://dx.doi.org/10.1016/j.worlddev.2025.107001}, DOI={10.1016/j.worlddev.2025.107001}, journal={World Development}, publisher={Elsevier BV}, author={Erlich, Aaron and Gans-Morse, Jordan and Nichter, Simeon and Holverscheid, Arne}, year={2025}, month=Oct, pages={107001} }
| 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 |