Primary Outcomes (explanation)
- Moral Universalism. We measure universalism taking inspiration from Cappelen et al. (2025) and Enke et al. (2024). Specifically, we ask participants a hypothetical question. "Suppose you have earned 1,000 EUR, but must give it away to two other people. You cannot keep any of the money to yourself. In all decisions, assume that both people have the same everyday standard of living. You can allocate the money between both people any way you wish?". The same question is repeated for several ingroup-members such as: (i) their family, (ii) a friend, (iii) a neighbour, (iv) someone who shares their religious beliefs, or (v) ethnic background, and (vi) a random stranger from their country.
To measure the effect of inequality on moral universalism, we construct an index by averaging participants’ allocations across all in-group categories and then inverting the resulting score. The final index ranges from 0 to 1000, where higher values indicate greater moral universalism and lower values indicate stronger in-group favouritism. The reason for creating an index is that the six variables are correlated with one another; treating them as independent hypotheses would overstate the effective number of tests, as they capture closely related dimensions of a common moral orientation. Using the index helps to increase precision and reduce noise by reducing the risk of spurious inference due to multiple testing.
- Petition. As a behavioral outcome, participants are given the opportunity to sign a petition advocating for increased action against inequality. The outcome is a binary indicator equal to one if the participant chooses to sign the petition and zero otherwise. This design allows us to examine whether exposure to inequality differentially affects support for policies targeting distinct domains of inequality.
- Volunteering exercise. Lasty, participants are invited to a volunteering exercise, where they can spend additional time (not be remunerated), to perform an effort task with the purpose of gaining additional money to support the activity of a well-established charity that fights to reduce inequality in all its forms. Specifically, they are presented with a series of small tasks. In each task, they are asked to select a specific number using a slider. For each page they completed accurately, they donate 0.1 EUR to the charity selected. They can complete up to 3 tasks in total, and they are asked before each one whether they would like to continue. Participation is completely voluntary, and they can stop at any time by clicking "No". We consider two alternative outcomes: a binary variable equal to one if the participant agrees to participate, and zero otherwise; and a continuous variable capturing performance in the task.