Construction of alternative hypotheses for randomization tests with ordinal outcomes

TL;DR

This paper develops a method to construct sequences of alternative hypotheses for ordinal outcomes, enabling better assessment of the power of randomization tests in treatment-control experiments.

Contribution

It introduces a novel approach to create alternative hypotheses for ordinal data, enhancing the evaluation of randomization test effectiveness.

Findings

Sequences of alternative hypotheses increase with departure from null.

Method improves power analysis of randomization tests.

Applicable to treatment-control experiments with ordinal outcomes.

Abstract

For ordinal outcomes, we construct sequences of alternative hypotheses in increasing departures from the sharp null hypothesis of zero treatment effect on each experimental unit, to help assess the powers of randomization tests in randomized treatment-control experiments.