Pseudoreplication is a pseudoproblem.

Abstract

Pseudoreplication is one of the most influential methodological issues in ecological and animal behavior research today. At its inception, the idea of pseudoreplication highlighted important concerns about the design and analysis of experiments in ecology. The doctrine purported to provide a unified view of experimental design and analysis, wherein precise criteria could be used to assess manuscripts and research proposals for acceptance or rejection. Few methodological doctrines have had as much impact as pseudoreplication, yet there has been very little critical analysis of it. In this paper, the authors extend the growing criticism of the concept of pseudoreplication. The authors argue that the core ideas behind pseudoreplication are based on a misunderstanding of statistical independence, the nature of control groups in science, and contexts of statistical inference. The authors also highlight how other areas of research have found and responded to similar issues in the design and analysis of experiments through the use of more advanced statistical methods. Ultimately, there are no universal criteria for accepting or rejecting experimental research; all research must be judged on its own merits.

Male vocal competition is dynamic and strongly affected by social contexts in music frogs

“…1). Though pseudoreplication is still being debated in ecological and animal behavior studies (Hurlbert 1984;McGregor et al 1992;Searcy et al 1997;McGregor 2000;Kroodsma et al 2001;Oksanen 2001;Hurlbert 2004;Oksanen 2004;Freeberg and Lucas 2009;Schank and Koehnle 2009), we controlled for the possible effects of pseudoreplication on our conclusions in the present study. Both the HSA and LSA calls used in the present study contained five notes recorded from the same individual with both temporal and frequency parameters close to average values for the population.…”

“…If the values were not normally distributed, they were transformed to square roots because the data were positively skewed (Munro 2005). Although pseudoreplication may have affected the data in the present experiments, it was still possible to evaluate differences in responsive calling across conditions using appropriately designed statistical tests based on ANOVAs (Hurlbert 2004;Schank and Koehnle 2009). Accordingly, two-way within-subject ANOVA (i.e., 2-way repeated measures ANOVA) was employed with the factors of ''acoustic stimulus'' and ''phase/panel'' for Experiment 1 and with the factors of ''acoustic stimulus/timing of call'' and ''ISI'' for experiment 2.…”

How will land use affect air temperature in the surface boundary layer? Lessons learned from a comparative study on the energy balance of an oak savanna and annual grassland in California, USA

“…In recent years, there has been a growing body of literature defending large-scale studies (Carpenter et al, 1995;Oksanen, 2001;Schank and Koehnle, 2009). Oksanen (2001), for example, concludes that 'The concept pseudoreplication amounts to entirely unwarranted stigmatization of a reasonable way to test predictions referring to largescale systems'.…”

Response of macrophyte communities to flow regulation in mountain streams

“…Probability levels were calculated using Monte Carlo simulations, based on 99,999 random reassignments of values within each up/downstream pair (Wilcoxon paired test) or 10,000 reassignments to the two upstream/ downstream groups (Mann-Whitney test). We used this Monte Carlo approach to deal with the possible non-independence of sites belonging to the same river/watershed (Schank and Koehnle 2009).…”