publish results selectively on the basis of outcomes. We agree with W&H that scientific publication is associated with costs in terms of writing and publishing [14], which should be diminished. However, we disagree with their notion that too many published results lead to what others have coined the “cluttered office phenomenon”, meaning that in an office full of academic papers it is hard to tell the good ones from the bad ones [15]. The reason is that statistical methods like meta-analysis have been developed to synthesize all the available evidence to avoid such a clutter of diverging results. Given the potentially far-reaching implications of W&H's conclusions for the publishing of empirical results in many scientific fields, we here criticize their conclusions. Specifically, we reanalyze their simulation scenario and results, and expand their simulations by including a scenario of a null effect (i.e., with false positive findings).
