Home

Computational Ecology and Software, 2011, 1(4): 218-223
[XML] [EndNote Tagged] [BibTex] [ Full PDF (50K)] [Comment Article]

Article

Estimating survival rates in ecological studies with small unbalanced sample sizes: an alternative Bayesian point estimator

Christian Damgaard1 , Adeline Fayolle2
1Department of Bioscience, Aarhus University, Vejlsovej 25. 8600 Silkeborg, Denmark
2Cirad, Environments and Societies Department, 'Goods and Services of Forest Ecosystems' Research Unit, Campus International de Baillarguet, TA 10C, BP 5035, Montpellier, 34035, France

Received 18 August 2011; Accepted 23 September 2011; Published online 1 December 2011
IAEES

Abstract
Increasingly, the survival rates in experimental ecology are presented using odds ratios or log response ratios, but the use of ratio metrics has a problem when all the individuals have either died or survived in only one replicate. In the empirical ecological literature, the problem often has been ignored or circumvented by different, more or less ad hoc approaches. Here, it is argued that the best summary statistic for communicating ecological results of frequency data in studies with small unbalanced samples may be the mean of the posterior distribution of the survival rate. The developed approach may be particularly useful when effect size indexes, such as odds ratios, are needed to compare frequency data between treatments, sites or studies.

Keywords frequency data; binomial distribution; small sample size; unbalanced design; effect size index; log response ratio.


International Academy of Ecology and Environmental Sciences. E-mail: office@iaees.org
Copyright © 2009-2024 International Academy of Ecology and Environmental Sciences. All rights reserved.
Web administrator: office@iaees.org, website@iaees.org; Last modified: 2024/4/26


Translate page to: