Comparer des méthodes

	Expérience de laboratoire bloquée	Schéma en carré latin et en carré gréco-latin
Domaine	Plans d'expériences	Plans d'expériences
Famille	Process / pipeline	Hypothesis test
Année d'origine	1926–1935	1935
Auteur d'origine	Ronald A. Fisher	Ronald A. Fisher
Type	Controlled experimental design with blocking	Parametric blocked ANOVA
Source fondatrice	Fisher, R. A. (1935). The Design of Experiments. Oliver and Boyd.	Montgomery, D. C. (2017). Design and Analysis of Experiments (9th ed.). Wiley. ISBN: 978-1119492443
Alias	blocked lab experiment, laboratory randomized block design, RBD laboratory study, blocked within-lab experiment	Latin Square, Greco-Latin Square, Latin Kare ve Greco-Latin Kare Deseni
Apparentées	5	5
Résumé	A blocked laboratory experiment is a controlled laboratory study in which experimental units are grouped into homogeneous blocks before treatment assignment, and treatments are then randomly assigned within each block. Blocking removes the influence of a known nuisance variable — such as participant batch, equipment run, or testing day — from the error term, increasing the precision of treatment comparisons without expanding sample size.	The Latin square design is a blocked experimental design that simultaneously controls two independent nuisance factors — the row block and the column block — so that each treatment appears exactly once in every row and every column of an n×n arrangement. Formalised by Ronald A. Fisher in his 1935 monograph The Design of Experiments, the design dramatically reduces experimental error by absorbing variation from two extraneous sources before the treatment effects are estimated.