Why We Need More Replication Studies To Keep Empirical Knowledge in Check