The identification of the causal effects of grade retention policies is of enormous relevance for researchers and policymakers alike. Taking advantage of the availability of more detailed longitudinal datasets, researchers have been able to apply different identification strategies that address the classical problems of selection bias and unobserved heterogeneity that have plagued previous studies on the effect of retention. We present a systematic literature review of empirical studies aiming to unveil the causal effects of retention. This study underlines the need to consider and evaluate different kinds of grade retention polices as their effects vary depending on several dimensions (such as timing of the policy, comparison groups, length of the effects or institutional settings). According to the results of our review, we conclude that grade retention is unlikely to be an efficient policy as the costs associated to the policy can easily outweigh the potential (weak) benefits of retention. It is therefore necessary to consider alternative policies to retention, or policies that can be used in combination with it, in order to enhance the performance of low achievers, in particular those students at risk characterized by a low ability profile.

中文翻译：

---

保留政策的效果：应用因果推论的实证研究的文献综述

成绩保留政策的因果关系识别对研究人员和政策制定者都具有重大意义。利用更详细的纵向数据集的可用性，研究人员已经能够应用不同的识别策略来解决选择偏见和未观察到的异质性的经典问题，这些问题困扰了先前有关保留效果的研究。我们提出了一项实证研究的系统文献综述，旨在揭示保留的因果关系。这项研究强调需要考虑和评估不同种类的等级保留策略，因为它们的效果取决于几个方面（例如政策的时间安排，比较小组，效果的持续时间或机构设置）。根据我们的审查结果，我们得出的结论是，保留等级不太可能是一种有效的策略，因为与该策略相关的成本很容易超过保留的潜在（弱）收益。因此，有必要考虑保留政策的替代政策，或者可以与保留政策结合使用的政策，以提高成绩不佳的学生，特别是那些能力低下的学生的表现。