A sequential selectivity model of the decisions of arbitrators

The study investigates arbitrator gender and grievant gender effects on the decisions of 146 arbitrators rendered on a hypothetical drug‐testing case. The effects are estimated using a sequential selectivity model. This approach takes into account the sequential and conditional nature of arbitrators' decisions. The results indicate that arbitrators are more likely to overturn terminations when the grievant is female, and that female arbitrators are less likely to fully reinstate grievants when the termination is overturned. Other results include the finding that older arbitrators treated grievants more harshly by more often awarding only partial reinstatements. When suspensions are given, however, we find that older arbitrators award shorter suspensions than do younger arbitrators. It is also found that more experienced arbitrators awarded longer suspensions than less experienced arbitrators.