Abstract
Replication experiments purport to independently validate claims from previous research or provide some diagnostic evidence about their reliability. In practice, this value of replication experiments is often taken for granted. Our research shows that in replication experiments, practice often does not live up to theory. Most replication experiments in practice are confounded and their results multiply determined, hence uninterpretable. These results can be driven by the true data generating mechanism, issues present in the original experiment, discrepancies between the original and the replication experiment, new issues introduced in the replication experiment, or combinations of any of these factors. The answers we are looking for with regard to the true state of nature require a rigorous and meticulous investigative process of eliminating errors and singling out elementary or pure cases. In this paper, we introduce the idea of a minimum viable experiment that needs to be identified in practice for replication results to be clearly interpretable. Most experiments are not replication-ready and before striving to replicate a given result, we need theoretical precision or systematic exploration to discover empirical regularities.