Comparer des méthodes
Examinez les méthodes sélectionnées côte à côte ; les lignes qui diffèrent sont mises en évidence.
| Expérimentation adaptative sur le terrain× | Expérience à bras multiples× | |
|---|---|---|
| Domaine | Plans d'expériences | Plans d'expériences |
| Famille | Process / pipeline | Process / pipeline |
| Année d'origine≠ | 1990s–2000s (formalized in field economics and development research contexts) | 1990s–2000s (clinical formalization); multi-arm concept implicit in ANOVA-era factorial designs |
| Auteur d'origine≠ | Developed at the intersection of adaptive trial methodology (Berry, Bauer) and field experimentation (Duflo, Kremer, List) | Developed within clinical trials methodology; formalized by Parmar, Royston and colleagues (UK MRC CTU, early 2000s) |
| Type≠ | Adaptive experimental design conducted in naturalistic settings | Experimental design |
| Source fondatrice≠ | Berry, D. A. (2004). Bayesian statistics and the efficiency and ethics of clinical trials. Statistical Science, 19(1), 175–187. DOI ↗ | Royston, P., Parmar, M. K. B., & Qian, W. (2003). Novel designs for multi-arm clinical trials with survival outcomes with an application in ovarian cancer. Statistics in Medicine, 22(14), 2239–2256. DOI ↗ |
| Alias | adaptive field trial, sequentially adaptive field experiment, responsive field experiment, adaptive randomized field study | multi-arm trial, multiple-arm experiment, multi-group experiment, many-arm design |
| Apparentées≠ | 6 | 5 |
| Résumé≠ | An adaptive field experiment is a randomized study conducted in a real-world environment in which pre-specified decision rules allow the researcher to modify the trial as interim data accumulate — for example, by reallocating participants toward more effective arms, adjusting sample size, or stopping early for efficacy or futility — all while maintaining statistical integrity. | A multi-arm experiment simultaneously compares three or more treatment or intervention conditions — each called an arm — against a shared control or against one another. By testing multiple alternatives in a single study, it yields more information per participant than running separate two-group experiments sequentially, while controlling the overall Type I error rate through pre-specified comparison strategies. |
| ScholarGateJeu de données ↗ |
|
|