Logistic regression results for the CONLOW and CONHIGH conditions.

The table shows the hierarchical four-level logistic regression estimates on whether (1) or not (0) a relation is present after an evaluation moment in the CONLOW and the CONHIGH conditions. Creation of relation is a dummy variable indicating whether the initial change was to create a relationship (1) or remove a relationship (0). Own outcome denotes the natural logarithm of the difference in own outcome between the situation with and without the relationship. Others' outcome denotes the natural logarithm of the difference in others' outcomes between the situation with and without the relationship. Equality denotes the difference in equality between the situation with and without the relationship. W1 * own outcome, W2 * others' outcomes, and W3 * equality denote the interaction effects between the aforementioned variables and the associated social preference measurements derived in the ring game. Both outcome variables and measures of social preference are centered around their respective means. Random effects are estimated at the level of the session, the decision maker, and the directed dyad under consideration. Numbers of sessions, decision makers, directed dyads, and decisions are given, as are the log likelihoods. P-values are in parentheses.