Table 3: Heterogeneous effects of treatments on donations
Notes: The table reports results from OLS regressions. The outcome variable in columns (1), (3), (5), and (7) is the amount donated to a targeted group of victims of violence (in cents); in columns (2), (4), (6), and (8), it is the share of donation in cash, conditional on donating. Ukraine and Africa are binary indicators (0-1) for the assignment to one of the treatment conditions, based on the nationality of the refugees. The control condition Local is omitted. College is a 0-1 indicator for the achievement of at least a college degree by a respondent; Right-wing score is a variable ranging from zero (leftmost leaning beliefs) to ten (rightmost orientation); High EU identity is an indicator equal to one if a respondent feels a strong or partial belonging to Europe and zero if they feel little or no belonging at all. Additional controls (not reported in the table) include an indicator for high perceived salience of immigration by the respondents, gender, and age. Robust standard errors are in parentheses, and FDR adjusted q-values are in brackets. ∗∗∗ significant at the 1% level, ∗∗ at the 5% level, ∗ at the 10% level.
Chart: CGOGet the dataEmbedDownload imageCreated with Datawrapper
