Pseudoreplication might often occured in the ecological researches during sampling, and there's also a saying that pseudoreplication happend during statistic analysis. So, I'm quite confused on this concept, and also wondering how to avoid pseudoreplication, could you please illustrate an example?
Hello!
http://www.mat.ufrgs.br/~giacomo/Softwares/R/Crawley/Crawley%20-%20The%20Book%20R/ch19.pdf
https://www.zoology.ubc.ca/biol548/lecturepdf/05.Models%20with%20random%20effects.pdf
Hello Jie Chen ,
Pseudo-replication can be defined as the use of replications that are not independent of each other. In experimentation, it is a mistake to use replications (of treatments) that are not independent of each other. That is, take replications of the same experimental unit and analyze them as if they were independent. Many do so to increase the sample size in order to increase the precision of the estimates, but for comparative purposes, the size of that sample should reflect the number of times the effect studied was examined independently. For example, this happens when repeated measurements are taken on the same set of subjects. In this case, the data taken are not independent and the averages per subject (experimental unit) or appropriate statistical models that take into account the repetition of the measurements in each experimental unit (example: mixed effects models) should be used.
In summary, it is better to use data from different experimental units so that there is independence between them. If you cannot avoid that, use the appropriate models for data analysis.
This link can also give you interesting elements: https://influentialpoints.com/Training/pseudoreplication.htm
Regards,
Mariel !
"Pseudoreplication is defined as the use of inferential statistics to test for treatment effects with data from experiments where either treatments are not replicated (though samples may be) or replicates are not statistically independent. [...] As a safeguard against both it and preexisting gradients, interspersion of treatments is argued to be an obligatory feature of good design. Especially in small experiments, adequate interspersion can sometimes be assured only by dispensing with strict randomization procedures." (Hurlbert, 1984)
In order to thoroughly understand pseudoreplication, its implications and impacts on experimental studies and how to prevent it in you sampling desing, I suggest you start from this article (Hurlbert, S.H. 1984. Pseudoreplication and the Design of Ecological Field Experiments. Ecological Monographs 54 (2), 187-211.Article Pseudoreplication and the Design of Ecological Field Experiments
). This is the "Bible" of pseudoreplication: you can then search for more recent literature on this topic (use the "Cited by" function in Google Scholar).Hope this helps.
L
- Badges
- Science topic
- Similar topics
- Medicine
- Physiology
- Biosignals
- Biomedical Signal Processing
- Statistical Analysis