Data from: Conflicting identities and cooperation between groups: Experimental evidence from a mentoring program
Data files
Jul 02, 2025 version files 34.90 KB
-
Espin_et_al_ProcB_dataset.csv
9.72 KB
-
Espin_et_al_ProcB_dataset.dta
15.45 KB
-
Espin_et_al_ProcB_script.do
6.86 KB
-
README.md
2.88 KB
Abstract
We experimentally investigate cooperation in 14 centres of a mentoring program where participants have two possible natural identities—individuals raised under legal guardianship, suffering a negative stereotype (G; n=112) and users without such a social stigma (NG; n=82). Participants played a Prisoners’ Dilemma game with an anonymous partner from the same centre (centre-ingroup) and from another centre (centre-outgroup). The folder contains the raw data in csv and dta (STATA) format and the script (STATA do file) used to define all the variables used and conduct all the analyses reported in the article. The analyses in the script appear in the exact order of appearance in the text, starting from the Methods section and then the Results section.
Associated article: Antonio M. Espín, María Paz Espinosa, María J. Vázquez-De Francisco, Pablo Brañas-Garza. Proceedings B. DOI: 10.1098/rspb.2025.1363
The folder contains the raw data in csv (Espin_et_al_ProcB_dataset.csv) and STATA (Espin_et_al_ProcB_dataset.dta) format, and the script (STATA do file: Espin_et_al_ProcB_script.do) used to define all the variables used and conduct all the analyses reported in the article. The analyses in the script appear in the exact order of appearance in the associated manuscript text, starting from the Methods section and then the Results section.
Variable description:
cod_center - code of the centre the participant belongs to
id - id code of the participant
id_incenter - id code of the participant within the centre
pop_propGcenter - proportion of G users (raised under legal guardianship) in the centre population based on administrative data
pair_enumerator - code of the pair of enumerators conducting the experiment
G - dummy variable, =1 if the participant is G (raised under legal guardianship)
NG - dummy variable, =1 if the participant is NG (not raised under legal guardianship)
age - participant age in years
female - dummy variable, =1 if the participant is a female
spanish - dummy variable, =1 if the participant is Spanish
history - number of weeks the participant had been in the program before the experiment
order_infirst - dummy variable, =1 if the participant played the prisoners' dilemma (PD) first with the centre-ingroup partner
coopin - dummy variable, =1 if the participant cooperates (chooses C) with the centre-ingroup
coopout - dummy variable, =1 if the participant cooperates (chooses C) with the centre-outgroup
Missing values: following the article's Methods section, the script curates the initial sample (n=216) in several steps until reaching the final sample (n=194) used for the statistical analyses, which does not contain missing values. The key part of the Methods section reads as follows:
This protocol resulted in a sample of 216 participants, with 127 G (58.8%) and 89 NG individuals (41.2%). For 14 participants, the date when they entered the program could not be confirmed. The other 7 participants failed to provide an unambiguous response for both PD games. Finally, in one centre, only one individual showed up, and therefore, their PD decision about the centre-ingroup had no real consequences. After excluding these 22 participants, the final sample consists of 194 individuals (112 Gs, or 57.7%, and 82 NGs, or 42.3%).
Human subjects data
We confirm that we received explicit consent from participants to publish the de-identified data in the public domain. All participants are identified by an anonymous numeric code in the data files.