方法对比

	试点自然实验	试点随机对照试验
领域	实验设计	实验设计
方法族	Process / pipeline	Process / pipeline
起源年份	2000s–2010s (as formalized practice)	1990s–2000s (methodological formalization)
提出者	Combination of natural experiment tradition (Dunning, Angrist, Pischke) and pilot study methodology	Formalized through clinical trials methodology community
类型	Quasi-experimental feasibility design	Experimental feasibility design
开创性文献	Dunning, T. (2012). Natural Experiments in the Social Sciences: A Design-Based Approach. Cambridge University Press. ISBN: 9781107017412	Thabane, L., Ma, J., Chu, R., Cheng, J., Ismaila, A., Rios, L. P., ... & Goldsmith, C. H. (2010). A tutorial on pilot studies: the what, why and how. BMC Medical Research Methodology, 10(1), 1.
别名	feasibility natural experiment, preliminary natural experiment, pilot quasi-experiment, exploratory natural experiment	pilot RCT, feasibility RCT, pilot trial, preliminary RCT
相关	4	5
摘要	A pilot natural experiment is a small-scale preliminary study that exploits an existing exogenous event or policy variation to test whether a full natural experiment is viable. It preserves the core logic of natural experiments — using real-world discontinuities to approximate causal inference — while explicitly scoping the work to assess data availability, group comparability, effect detectability, and procedural feasibility before committing resources to a larger study.	A pilot randomized controlled trial (pilot RCT) is a small-scale, fully randomized experiment conducted before a definitive RCT to test the feasibility of study procedures, estimate key parameters such as recruitment rates and effect-size variability, and identify practical barriers. It uses the same randomization, intervention, and measurement protocol as the planned full trial but on a fraction of the target sample. The goal is not to confirm efficacy but to refine and justify the main trial design.