Elsevier

Animal Behaviour

Volume 120, October 2016, Pages 223-234
Animal Behaviour

Special Issue: Conservation Behaviour
A simple statistical guide for the analysis of behaviour when data are constrained due to practical or ethical reasons

https://doi.org/10.1016/j.anbehav.2015.11.009Get rights and content

Highlights

  • I review statistical approaches for dealing with common constraints in animal behaviour studies.

  • I suggest analytical solutions to these constraints that outperform conventional methods.

  • I show how these statistical approaches can be used in the R statistical environment.

  • Conservationists as well as evolutionary ecologists should find these approaches useful.

Here, I provide a practical overview on some statistical approaches that are able to handle the constraints that frequently emerge in the study of animal behaviour. When collecting or analysing behavioural data, several sources of limitations, which can raise either uncertainties or biases in the parameter estimates, need to be considered. In particular, these can be issues about (1) limited sample size and missing data, (2) uncertainties about the identity of subjects and the dangers posed by pseudoreplication, (3) large measurement errors resulting from the use of indicator variables with nonperfect reliability or variables with low repeatability, (4) the confounding effect of the within-individual variation of behaviour and (5) phylogenetic nonindependence of data (e.g. when substitute species are used). I suggest some simple analytical solutions to these problems based on existing methodologies and on a consumable language to practitioners. I highlight how randomization and simulation routines, generalized linear mixed models, autocorrelation models, phylogenetic comparative methods and Bayesian statistics can be exploited to overcome the inefficient performance of some conventional statistical approaches with typical behavioural data. To enhance the accessibility of these methodologies, I demonstrate how they can be brought into practice in the R statistical environment, which offers flexible statistical designs. Although the primary motivation behind this discussion was to help animal behaviourists who address questions in relation to conservation, I also hope that researchers working on the evolutionary ecology of behaviour will also find some material useful.

Section snippets

Limited sample size

Limited sample size is one of the most obvious constraints that confronts animal behaviourists (Taborsky, 2010), especially when working on conservation-related issues (Bradshaw and Brook, 2010, Martinez-Abrain, 2014). For a variety of reasons that arise from the special characteristics of the studied species, in combination with the difficulty of assaying behaviours and ethical policies, it is impossible to acquire an ideal sample that would be representative of the real world. This is a

Noninvasive sampling methods: measurement error for surrogate variables

It is often required to use noninvasive methods for sampling when the model species is to be protected, but this is also becoming preferable practice in the modern study of animal behaviour for ethical reasons (Mench, 2000). Several methods are now available that permit the estimation of certain physiological traits (such as hormone levels, parasite load, heart rate, metabolic status, immunocompetence) through surrogate variables that help reduce or completely eliminate the stress that animals

Unknown identity of subjects

Given the ethical and practical concerns regarding the capture of rare or endangered species of animals, conservation biologists cannot always directly mark individuals, and in such cases they have to make behavioural observations without knowing the identity of the focal animals. Behaviour is a trait that, unlike many morphological or physiological traits, permits data collection from a distance (using binoculars or video cameras) to some extent. Therefore, if reliable means of identifying

Mixed Modelling: the Specific Hierarchical Structure of Behavioural Data

Studying the adaptive role of behaviour in the light of the rapidly and unpredictably changing environment may be interesting from a conservation aspect, as such adaptation processes are highly relevant under the recently occurring climatic changes. An individual's behaviour can change rapidly from moment to moment, and this plasticity offers a mean by which the animal can quickly react to an emerging stress factor in the environment (Dingemanse, Kazem, Réale, & Wright, 2010). Given that quick

Conclusions

Here I have provided an overview on various statistical methods that can be useful for conservation biologists and animal behaviourists when analysing behavioural data that are often loaded with various constraints. The common theme appearing in this discussion is that some aspects of these constraints should be regarded as sources of noise that generate uncertainty around the estimated parameters (e.g. small sample size, unknown identity of subjects). These uncertainties are inherent

Acknowledgments

I thank E. Fernández-Juricic and B. A. Schulte for inviting me to contribute to the Special Issue on Conservation Behaviour. I am also grateful to one anonymous referee and Matthew Symonds for their constructive comments on an earlier version of the manuscript. During this study, the author was supported by funds from the Spanish government within the frame of the Plan Nacional Programme (no. CGL2012- 38262 and no. CGL2012-40026-C02-01) and from the National Research, Development and Innovation

References (101)

  • R.J.G. Pereira et al.

    Seasonal changes in fecal testosterone concentrations and their relationship to the reproductive behaviour, antler cycle and grouping patterns in free-ranging male pampas deer (Ozotoceros bezoarticus bezoarticus)

    Theriogenology

    (2005)
  • M.V. van de Pol et al.

    A simple method for distinguishing within- versus between-subject effects using mixed models

    Animal Behaviour

    (2009)
  • P. Sprau et al.

    Song type sharing in common nightingales, Luscinia megarhynchos, and its implications for cultural evolution

    Animal Behaviour

    (2010)
  • P.A. Stephens et al.

    Inference in ecology and evolution

    Trends in Ecology & Evolution

    (2007)
  • P. Taberlet et al.

    Non-invasive genetic sampling and individual identification

    Biological Journal of the Linnean Society

    (1999)
  • J.H.G. Williams et al.

    Second to fourth finger ratio and possible precursors of developmental psychopathology in preschool children

    Early Human Development

    (2003)
  • S.C. Adolph et al.

    Estimating phenotypic correlations: correcting for bias due to intraindividual variability

    Functional Ecology

    (2007)
  • S.C. Adolph et al.

    Estimating maximum performance: effects of intraindividual variation

    Journal of Experimental Biology

    (2008)
  • L.M. Aplin et al.

    Experimentally induced innovations lead to persistent culture via conformity in wild birds

    Nature

    (2015)
  • A.C. Atkinson

    Plots, transformations and regression: An introduction to graphical methods of diagnostic regression analysis

    (1985)
  • A.J. Bishara et al.

    Reducing bias and error in the correlation coefficient due to nonnormality

    Educational and Psychological Measurement

    (2015)
  • B. Bolker

    Ecological models and data in R

    (2007)
  • K.A. Bollen

    Structural equations with latent variables

    (1989)
  • C.J.A. Bradshaw et al.

    The conservation biologist's toolbox: principles for the design and analysis of conservation studies

  • J.P. Buonaccorsi

    Measurement error: Models, methods, and applications

    (2010)
  • K.P. Burnham et al.

    Model selection and multimodel inference: A practical information-theoretic approach

    (2002)
  • K.S. Button et al.

    Power failure: why small sample size undermines the reliability of neuroscience

    Nature Reviews Neuroscience

    (2013)
  • T. Caro et al.

    Use of substitute species in conservation biology

    Conservation Biology

    (2005)
  • A. Chesher

    The effect of measurement error

    Biometrika

    (1991)
  • J. Cohen

    Statistical power analysis for the behavioural sciences

    (1988)
  • J. Cohen

    The earth is round (p< .05)

    American Psychologist

    (1994)
  • P. Congdon

    Bayesian statistical modelling

    (2006)
  • N.J. Dingemanse et al.

    Fitness consequences of avian personalities in a fluctuating environment

    Proceedings of the Royal Society B: Biological Sciences

    (2004)
  • N.J. Dingemanse et al.

    Quantifying individual variation in behaviour: mixed-effect modelling approaches

    Journal of Animal Ecology

    (2013)
  • N.J. Dingemanse et al.

    Defining behavioural syndromes and the role of ‘syndrome deviation’ in understanding their evolution

    Behavioral Ecology and Sociobiology

    (2012)
  • D.S. Dobkin

    Functional and evolutionary relationhips of vocal copying phenomena in birds

    Zeitschrift für Tierpsychologie

    (1979)
  • B. Efron et al.

    An introduction to the bootstrap. Monographs on Statistics and Applied Probability 57

    (1993)
  • J. Fox

    Applied regression analysis and generalized linear models

    (2008)
  • R.P. Freckleton

    Dealing with collinearity in behavioural and ecological data: model averaging and the problems of measurement error

    Behavioral Ecology and Sociobiology

    (2011)
  • W.A. Fuller

    Measurement error models

    (1987)
  • A. Gałecki et al.

    Linear mixed-effects models using R: A step-by-step approach

    (2013)
  • L.Z. Garamszegi

    Comparing effect sizes across variables: generalization without the need for Bonferroni correction

    Behavioral Ecology

    (2006)
  • L.Z. Garamszegi

    Modern phylogenetic comparative methods and their application in evolutionary biology: Concepts and practice

    (2014)
  • L.Z. Garamszegi et al.

    Changing philosophies and tools for statistical inferences in behavioural ecology

    Behavioral Ecology

    (2009)
  • L.Z. Garamszegi et al.

    Behavioural syndromes, syndrome deviation and the within- and between-individual components of phenotypic correlations: when reality does not meet statistics

    Behavioral Ecology and Sociobiology

    (2012)
  • L.Z. Garamszegi et al.

    A meta-analysis of correlated behaviours with implications for behavioural syndromes: mean effect size, publication bias, phylogenetic effects and the role of mediator variables

    Evolutionary Ecology

    (2012)
  • A. Gelman

    Working through some issues

    Significance

    (2015)
  • A. Gelman et al.

    Bayesian data analysis

    (1995)
  • R.L. Gorsuch et al.

    Correlation coefficients: mean bias and confidence interval distortions

    Journal of Methods and Measurement in the Social Sciences

    (2010)
  • T.F. Hansen et al.

    Interpreting the evolutionary regression: the interplay between observational and biological errors in phylogenetic comparative studies

    Systematic Biology

    (2012)
  • Cited by (40)

    • Carry-over effects of pre-breeding diets on seahorse (Hippocampus reidi) reproductive success

      2021, Aquaculture
      Citation Excerpt :

      To my knowledge, there is not any seahorse rearing facility nor such a huge amount of seahorse breeders available worldwide. On the other hand, there is literature available that justify the use of pseudoreplicates in some circumstances and recommends that pseudoreplication not be invoked when reviewing research manuscripts and research proposals (Schank and Koehnle, 2009; Davies and Grey, 2015; Garamszegi, 2015). Statistical analyses were conducted in R v.3.6.1 (R Core Team, 2014).

    View all citing articles on Scopus
    View full text