Implications of interaction effects on the interpretation of factorial experiments in biology
Abstract
Factorial experiments are commonly used in biology, particularly to study the way in which two or more factors are related. The presence of a significant interaction indicates that the factors are not independent. In this case, two related aspects should be considered: 1) the post hoc analysis after detecting the significant interaction, and 2) the conclusion on the other hypotheses included in the model. The objective of this work is to discuss the importance and implicances of the significant interactions on the inferences of the hypotheses of factorial models with fixed effects. We include four hypothetical problems in ecology with hypothetical data reflecting different scenarios with significant interactions. In each example, we discuss the possible biological significance of the marginal means of one factor averaged over the levels of the other in the presence of significant interactions, their post hoc analysis, and inferences.
References
HURLBERT, SH. 1984. Pseudoreplication and the design of ecological field experiments. Ecol. Monogr. 54:187-211.
MILLIKEN, GA & DE JOHNSON. 1992. Analysis of messy data. Volume I: designed experiments. Chapman & Hall, London. 473 pp.
MONTGOMERY, DC. 1981. Design and analysis of experiments. J. Wiley & Sons, New York. 538 pp.
NETER, J; MH KUTTNER; CJ NACHTSHEIM & W WASSERMAN. 1996. Applied linear statistical models. 4ta edn. Irwin, Chicago. 1408 pp.
OTT, L. 1993. An introduction to statistical methods and data analysis. 4ta edn. Wadsworth. 132 pp.
SEARLE, SR. 1971. Linear models. J. Wiley & Sons, New York. 532 pp.
SHAW, RG & T MITCHELL-OLDS. 1993. ANOVA for unbalanced data: an overview. Ecology 74:1638-1645.
SOKAL, RR & J ROHLF. 1981. The principles and practice of statistics in biological research. 2da edn. W. H. Freeman, New York. 859 pp.
STEEL, RGD & JH TORRIE. 1988. Bioestadística. Principios y procedimientos. 2da edn. Mc Graw-Hill, México. 622 pp.
UNDERWOOD, AJ. 1997. Experiments in ecology. Their logical design and interpretation using analysis of variance. Cambridge University Press, Cambridge. 504 pp.
ZAR, JH. 1996. Biostatistical analysis. 3ra edn. Prentice Hall, Upper Saddle River. 662 pp.
Downloads
Published
How to Cite
Issue
Section
License
Authors retain their rights as follows: 1) by granting the journal the right to its first publication, and 2) by registering the published article with a Creative Commons Attribution License (CC-BY 4.0), which allows authors and third parties to view and use it as long as they clearly mention its origin (citation or reference, including authorship and first publication in this journal). Authors can make other non-exclusive distribution agreements as long as they clearly indicate their origin and are encouraged to widely share and disseminate the published version of their work.