手法を比較

	適応的実施要因実験	適応実験
分野	実験計画法	実験計画法
系統	Process / pipeline	Process / pipeline
提唱年	1950s (factorial foundations); adaptive extensions prominent from 1990s onward	1940s–1970s (sequential foundations); formalised in clinical and behavioural research by 1980s–2000s
提唱者	Rooted in Box & Hunter factorial design tradition; adaptive extensions formalised by Atkinson, Donev and others in optimal design theory	Abraham Wald (sequential analysis foundation); expanded by Robbins, Armitage, and others
種類	Experimental design	Experimental research design
原典	Atkinson, A., Donev, A., & Tobias, R. (2007). Optimum Experimental Designs, with SAS. Oxford University Press. ISBN: 978-0199296606	Chow, S. C., & Chang, M. (2008). Adaptive Design Methods in Clinical Trials. Chapman and Hall/CRC. ISBN: 978-1584886761
別名	adaptive full-factorial design, sequential full factorial experiment, adaptive complete factorial design, dynamic full factorial trial	adaptive design, response-adaptive randomization, adaptive trial, adaptive randomization
関連	5	5
概要	An adaptive full factorial experiment is an experimental design that starts with a complete crossing of all factors and all their levels, then uses interim data to modify subsequent runs — dropping unpromising factor levels, adding new ones, or re-allocating replication — while preserving the full factorial structure within each phase. This integration of full factorial coverage with adaptive decision rules allows researchers to explore all main effects and interactions without committing to a fixed, inefficient run plan before any data are observed.	An adaptive experiment is an experimental design in which pre-specified rules allow the protocol to be modified — such as reallocating participants to better-performing arms, stopping early for efficacy or futility, or changing sample size — based on accumulating interim data, while maintaining statistical validity. Adaptive designs are widely used in clinical trials, behavioural economics, and online platform testing to improve efficiency and ethics without sacrificing inferential rigour.