Classical automated test assembly (ATA) methods assume fixed and known coefficients for the constraints and the objective function. This hypothesis is not true for the estimates of item response theory parameters, which are crucial elements in test assembly classical models. To account for uncertainty in ATA, we propose a chance-constrained…

In cluster-randomized trials (CRTs) of social work interventions, groups are assigned to treatment conditions. Conducting a power analysis ensures that enough groups are sampled for testing hypotheses. The power analysis method and inputs must be informed by the hypotheses, effects to be tested, and the data analysis plans. The authors present a…