Alternatives to Randomized Experiments
Alternatives to Randomized Experiments
复制标题
DOI：
10.1111/j.1467-8721.2009.01656.x
复制
发表时间：
2009-10-01
影响因子：
7.2
通讯作者：
West, Stephen G.
中科院分区：
文献类型：
--
作者：
West, Stephen G.
当随机实验可以实施并且其假设得到满足时，随机实验更适合于对因果关系进行推断。然而，假设可能会失败（例如，自然减员、治疗不依从）或随机化可能是不道德或不可行的。我描述了替代设计和统计方法，允许测试因果假设和目前的经验证据相关的替代设计。替代设计允许回答更广泛的研究问题，并允许更直接地概括因果效应;然而，当使用此类设计时，对因果效应大小的估计可能更加不确定。
Randomized experiments are preferred for making inferences about causality when they can be implemented and their assumptions are met. Yet assumptions can fail (e.g., attrition, treatment noncompliance) or randomization may be unethical or infeasible. I describe alternative design and statistical approaches that permit testing causal hypotheses and present current empirical evidence related to alternative designs. Alternative designs permit a wider range of research questions to be answered and permit more direct generalization of causal effects; however, when using such designs, estimates of the magnitude of the causal effect may be more uncertain.