Norms on in-group favoritisim across social contexts
How to cite
Marín, Isabel Rodríguez; Miller, Luis; Lane, Tom. 2024. "Norms on in-group favoritisim across social contexts." AEA RCT Registry. AEARCTR-0013896. https://doi.org/10.1257/rct.13896-1.0
Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.
* Primary investigator
Abstract
In a previous experimental study, we documented that social norms in discrimination games vary by social context. In particular, we show that participants rate in-group favouritism as more or less socially appropriate based on the social identity that is in play. Participants perceived it to be more socially inappropriate to discriminate on the basis of religious affiliation than on the basis of political identity. The study included also a behavioural task that showed an alignment between social norms and behaviours. Participants opted for decisions that favoured their in-groups more frequently in the social contexts where those decisions were regarded as more socially appropriate. This study was carried out in experimental laboratories in the UK and Spain between 2022 and 2023.
Here, we present a vignette experiment to measure social norms on in-group favouritism and group behaviour across a wide range of social identities. Our hypothesis is that participants will rate in-group favouritism differently in different contexts of social identity. We will be able to rank identities based on the level of in-group favouritism considered appropriate in each context. That ranking will be influenced by the agency that individuals grant to other people over features of their identity.
Intervention
The survey experiment consists of a vignette that briefly describes a ‘real situation from a previous study’. The situation described is the allocation task from previous behavioural experiments (AEARCTR-0010172/0011275/0013361), in which one subject has to decide how to allocate money between another in-group participant and an out-group participant. Then, subjects answering the survey are asked to rate the possible decisions that the participants in the behavioural tasks could make as: "Very socially appropriate", "Somewhat socially appropriate", "Somewhat socially inappropriate" and "Very socially inappropriate", considering not only their opinion but the social average. This question is not incentivised.
Experimental design
The dimension of identity that is primed in every treatment is varied exogenously. We will ask participants to rate the decisions of another participant identified with one social group, with the goal of making salient a particular social context. The dimension of identity that is made salient is varied exogenously between partisanship, religious affiliation, football team, gender, educational level and immigration status. This variation constitutes our main treatment. Along these treatment dimensions, characters in the vignette identified with two opposing sides: PP vs. PSOE, catholic vs. non-believer, Real Madrid vs. Barça, male vs. female, lower education vs. upper education, Spanish vs immigrant. Within each treatment, we will also randomly assign the allocator to one of the two identity groups. We will elicit own opinions and identification of individuals to account for their natural identities when examining their responses to the vignette norm-elicitation task.
Design details
Cited by
No citation data available for this registration yet.