2,245 rated or chosen profiles from 190 respondents, in up to 12 tasks of 1 profile varying on 9 attributes.

About this table

Description	190 employers and recruiters in Oman each rated 12 text resumes of graduate applicants (one deck of a 144-vignette fractional design, one occupation per respondent) varying nine attributes (gender, nationality, experience, activities, place of study, education, field, GPA, referral); outcomes: hire likelihood and trainability, 0-100.
Reference	Al Hatmi, K. (2026). Social networks, gender, and graduate hiring in a rentier state: Disentangling referral types and ascriptive signals in Omani labour markets. Research in Social Stratification and Mobility.
Licence	CC0 1.0
Source data	https://doi.org/10.7910/DVN/R2W8AW

Notes

Conjoint table (irw_conjoint_2). Single-profile text resumes (nine attributes) rated by Omani employers: rating_hire (likelihood of hiring) and rating_train (likelihood of being easy to train), both 0-100 as stored (the README says 0-10; anchors unknown). Fixed blocked design: each respondent saw the 12 vignettes of one of 12 decks (trial_deck) for one occupation (trial_job_type); no opt-out. Only a README, not the article, identifies the study.

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

Design

Topic	labour hiring
A profile represents	person
Country	OM
Language shown	en
Language of stored levels	en
Randomization restrictions	yes (stated by the source): fixed fractional design: only 144 designed vignettes (SAS D-efficient fractional factorial), 12 decks of 12; all respondents of a deck saw the same vignettes
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:

rating_hire	How likely is it that you would hire this applicant? [rating, 0 (unknown) to 100 (unknown)]
rating_train	If this applicant is hired, how likely is it that the applicant would be easy to train to meet the job demands? (wording from the design document) [rating, 0 (unknown) to 100 (unknown)]

Columns

attr_educationattr_experienceattr_extracurricularattr_fieldattr_genderattr_gradeattr_nationalityattr_place_of_studyattr_referralcov_age_groupcov_educationcov_gendercov_nationalityidprofilerating_hirerating_traintasktrial_decktrial_job_type

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("alhatmi_2026_oman_graduate_hiring", source = "conj")

# Python
pip install irw

import irw
df = irw.fetch("alhatmi_2026_oman_graduate_hiring", source="conj")

How to cite

@misc{https://doi.org/10.7910/dvn/r2w8aw,
  doi = {10.7910/DVN/R2W8AW},
  url = {https://dataverse.harvard.edu/citation?persistentId=doi:10.7910/DVN/R2W8AW},
  author = {Al Hatmi, Khalifa},
  keywords = {Social Sciences},
  title = {A dataset of employers’ hiring decision factors in Oman, examining what organizations look for when recruiting graduates and the underlying reasons},
  publisher = {Harvard Dataverse},
  year = {2026}
}

Respondents	190
Rows (one per respondent, task and profile)	2,245
Tasks per respondent (most)	12
Profiles per task	1
Attributes	9

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

alhatmi_2026_oman_graduate_hiring