13,120 rated or chosen profiles from 410 respondents, in up to 16 tasks of 2 profiles varying on 5 attributes.

About this table

Description	Discrete choice experiment on valence framing: 410 UK online respondents chose between two unlabelled train alternatives (cost, time, seat availability, service quality, CO2 emissions) whose attributes were worded in gain or loss frames; 16 forced choices each, eight framing versions.
Reference	Barnes, O., Hess, S., & Hancock, T. (2025). Valence framing in multi-attribute decision-making [Dataset]. University of Leeds.
DOI	10.31234/osf.io/3xh6t
Licence	CC BY 4.0
Source data	https://doi.org/10.5518/1709

Notes

Conjoint table (irw_conjoint_2). Forced choice between two unlabelled trains ("Which would you choose?") described by cost, travel time, seat availability, service quality and CO2 emissions, with the wording framed as gains or losses (seats free/taken, satisfied/dissatisfied customers, CO2 saved/missed); no opt-out. Each respondent saw 4 of 8 framing versions (trial_version, trial_frame_*), 4 tasks each. Level text is the displayed wording from the dictionaries; seat availability is a count in one train and a chance in the other, the relative CO2 statement sits in one train's column only (the other is (not shown)). Task order is inferred from the stored version order. 410 respondents.

These notes describe the source data, which the IRW reproduces as released. All notes are listed on the Data Notes page.

Design

Topic	climate energy environment
A profile represents	product service
Country	GB
Language shown	en
Language of stored levels	en
Randomization restrictions	observed in the data (not documented by the source): seat availability is stated as a count in one alternative and as a chance in the other; the relative CO2 statement appears in one alternative's column only; each respondent sees 4 of 8 framing versions (blocked Ngene-type design, 12 blocks)
Task order	inferred from row order
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. Task order or profile position was rebuilt from row order; analyses of position or task order should leave it out.

Outcomes:

choice	Which would you choose? [choice, forced]

Columns

attr_co2_emissionsattr_seat_availabilityattr_service_qualityattr_travel_costattr_travel_timechoicecov_age_groupcov_duration_seccov_gendercov_income_bandcov_risk_willingnesscov_speeder_flagidprofiletasktrial_frame_co2trial_frame_qualitytrial_frame_seatstrial_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.

Get the data

Download CSVno account needed Browse on Redivisexplore and query Croissant metadataHugging Face, Kaggle, OpenML

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("barnes_2025_valence_framing", source = "conj")

# Python
pip install irw

import irw
df = irw.fetch("barnes_2025_valence_framing", source="conj")

How to cite

@dataset{Barnes2025,
  author = {Barnes, O. and Hess, S. and Hancock, T.},
  title = {Valence framing in multi-attribute decision-making},
  year = {2025},
  publisher = {University of Leeds},
  doi = {10.5518/1709},
  url = {https://doi.org/10.5518/1709}
}

Respondents	410
Rows (one per respondent, task and profile)	13,120
Tasks per respondent (most)	16
Profiles per task	2
Attributes	5
Tasks with no profile chosen	0

IRW version	v584
Redivis dataset	irw_conjoint_2 v1.4
Redivis dataset DOI	10.57761/hk97-wv30
Manifest pin for this IRW version	v1.4
Metadata source	irw_meta v44.0

barnes_2025_valence_framing