Comparer des méthodes

	Expérience de terrain à simple aveugle	Expérience de terrain en double aveugle
Domaine	Plans d'expériences	Plans d'expériences
Famille	Process / pipeline	Process / pipeline
Année d'origine	Mid-20th century (blinding conventions formalised 1940s–1960s)	1960s onward (field experiment tradition); double-blind controls applied from 1970s in social and policy field trials
Auteur d'origine	Established practice in experimental social science and clinical research; codified by Campbell & Stanley (1963) and Shadish, Cook & Campbell (2002)	Fisher, R. A. (randomized field trials); double-blind practice traced to 19th-century clinical research, formalized for field settings by Campbell & Stanley (1963)
Type	Controlled field experiment with partial blinding	Experimental design
Source fondatrice	Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and Quasi-Experimental Designs for Generalized Causal Inference. Houghton Mifflin. ISBN: 978-0395615560	Gerber, A. S., & Green, D. P. (2012). Field Experiments: Design, Analysis, and Interpretation. W. W. Norton. ISBN: 978-0393979954
Alias	single-masked field experiment, field experiment with single blinding, single-blind natural-setting trial	double-masked field trial, double-blind naturalistic experiment, blinded field study, DB field experiment
Apparentées	6	5
Résumé	A single-blind field experiment combines real-world experimental conditions with partial blinding: either participants or outcome assessors — but not both — are kept unaware of treatment assignment. This design reduces demand characteristics or observer bias while preserving ecological validity, making it a practical middle ground when full double-blinding is logistically infeasible in naturalistic settings.	A double-blind field experiment combines the high external validity of a real-world field setting with double-blind masking, in which neither the participants nor the personnel delivering the treatment know who has been assigned to the treatment or control condition. This design controls simultaneously for participant expectation effects and for experimenter/enumerator demand effects, making it one of the most rigorous tools available for causal inference outside the laboratory.