Experiment: Conceptual Basis
Joao Gilberto Correa da Silva *
Department of Mathematics and Statistics, Federal University of Pelotas, Pelotas, RS, Brazil.
*Author to whom correspondence should be addressed.
Abstract
Texts and teaching of Experimental Statistics emphasize the statistical analysis of experiments and superficially consider the conceptual basis of experimental research. Definitions of basic concepts are imprecise, incoherent and ambiguous. This is the case, for example, with the concepts of experimental material, experimental factor, experimental unit and experimental error. In particular, the lack of distinction between experimental factor and unit factor and between the two classes of experimental factor: treatment factor and intrinsic factor leads to flaws in the plan and analysis of experiments that originate biased inferences. This approach gives rise to ignorance and misunderstanding of these concepts, underestimation of the importance of the planning of the experiment and, consequently, the inefficiency of many research. This paper reviews and reformulates important concepts with the purpose of contributing a rational basis for experimental research and, in particular, for Experimental Statistics. These revised concepts provides a basis for the formulation of an experiment structure that provides appropriate inferences for the achievement of the objectives of the experiment.
Keywords: Experimental statistics, experimental material, experimental factor, experimental unit, unit factor, experimental error