Applied statistics in ecology: common pitfalls and simple solutions

Steel, E. Ashley; Kennedy, Maureen C.; Cunningham, Patrick G.; Stanovick, John S.

doi:10.1890/es13-00160.1

Cited by 63 publications

(49 citation statements)

References 42 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In these examples, the quadrat, point and tree are the units of observation, whereas the site, transect or forest stand are the units of inference. Such ‘pseudoreplication’ is common in ecological data (Hurlbert, ; Steel, Kennedy, Cunningham, & Stanovick, ) and may be especially prevalent in observational data. Additionally, observational units may not be uniquely labelled (e.g., if quadrats are not marked with a permanent pin, volunteers do not stop at the same exact locations each time they walk a transect, or individual trees are not given a permanent identification tag).…”

Section: Methodsmentioning

confidence: 99%

Joint species distribution modelling for spatio‐temporal occurrence and ordinal abundance data

Schliep

Lany

Zarnetske

et al. 2017

Global Ecol Biogeogr

View full text Add to dashboard Cite

Aim: Species distribution models are important tools used to study the distribution and abundance of organisms relative to abiotic variables. Dynamic local interactions among species in a community can affect abundance. The abundance of a single species may not be at equilibrium with the environment for spreading invasive species and species that are range shifting because of climate change. Innovation:We develop methods for incorporating temporal processes into a spatial joint species distribution model for presence/absence and ordinal abundance data. We model non-equilibrium conditions via a temporal random effect and temporal dynamics with a vectorautoregressive process allowing for intra-and interspecific dependence between co-occurring species. The autoregressive term captures how the abundance of each species can enhance or inhibit its own subsequent abundance or the subsequent abundance of other species in the community and is well suited for a 'community modules' approach of strongly interacting species within a food web. R code is provided for fitting multispecies models within a Bayesian framework for ordinal data with any number of locations, time points, covariates and ordinal categories.Main conclusions: We model ordinal abundance data of two invasive insects (hemlock woolly adelgid and elongate hemlock scale) that share a host tree and were undergoing northwards range expansion in the eastern U.S.A. during the period 1997-2011. Accounting for range expansion and high inter-annual variability in abundance led to improved estimation of the species-environment relationships. We would have erroneously concluded that winter temperatures did not affect scale abundance had we not accounted for the range expansion of scale. The autoregressive component revealed weak evidence for commensalism, in which adelgid may have predisposed hemlock stands for subsequent infestation by scale. Residual spatial dependence indicated that an unmeasured variable additionally affected scale abundance. Our robust modelling approach could provide similar insights for other community modules of co-occurring species.biotic interactions, coregionalization, invasive species, Markov chain Monte Carlo, rank probability scores, vector autoregression | I NTR OD U CTI ONSpecies distribution models are commonly used in basic and applied ecological research to study the factors that define the distribution and abundance of organisms. They are used to quantify species' relationships with abiotic conditions, to predict species' response to land-use and climatic change and to identify potential conservation areas (Guisan & Zimmermann, 2000). Traditionally, species distribution models correlate static observations of the occurrence (presence and absence) or abundance of a species with abiotic variables. Occurrence and abundance of a species, however, can also change through time and across space through colonization and may not be at equilibrium

show abstract

Section: Methodsmentioning

confidence: 99%

Joint species distribution modelling for spatio‐temporal occurrence and ordinal abundance data

Schliep

Lany

Zarnetske

et al. 2017

Global Ecol Biogeogr

View full text Add to dashboard Cite

show abstract

“…GLMs function well on count data that include zeros (Bolker et al, 2008;McCullagh & Nelder, 1989); however, when there is an abundance of zeros relative to a Poisson or negative binomial error model, a hurdle or zero-inflated model is recommended when fitting models and interpreting the estimates and trends (Lambert, 1992;Mullahy, 1986). The increasing availability of software and accessible texts (e.g., Dunteman & Ho, 2006;Hoffmann, 2004), combined with the advantages of avoiding transformation of data and back-transformation of parameter estimates, has resulted in repeated recommendations to replace data transformation with GLM/ GLMM procedures (Lo & Andrews, 2015;O'Hara & Kotze, 2010;Steel, Kennedy, Cunningham, & Stanovick, 2013;Warton & Hui, 2011;Wilson & Hardy, 2002).…”

Section: Introductionmentioning

confidence: 99%

Count data in biology—Data transformation or model reformation?

St‐Pierre

Shikon

Schneider

2018

Ecology and Evolution

View full text Add to dashboard Cite

Statistical analyses are an integral component of scientific research, and for decades, biologists have applied transformations to data to meet the normal error assumptions for F and t tests. Over the years, there has been a movement from data transformation toward model reformation—the use of non‐normal error structures within the framework of the generalized linear model (GLM). The principal advantage of model reformation is that parameters are estimated on the original, rather than the transformed scale. However, data transformation has been shown to give better control over type I error, for simulated data with known error structures. We conducted a literature review of statistical textbooks directed toward biologists and of journal articles published in the primary literature to determine temporal trends in both the text recommendations and the practice in the refereed literature over the past 35 years. In this review, a trend of increasing use of reformation in the primary literature was evident, moving from no use of reformation before 1996 to >50% of the articles reviewed applying GLM after 2006. However, no such trend was observed in the recommendations in statistical textbooks. We then undertook 12 analyses based on published datasets in which we compared the type I error estimates, residual plot diagnostics, and coefficients yielded by analyses using square root transformations, log transformations, and the GLM. All analyses yielded acceptable residual versus fit plots and had similar p‐values within each analysis, but as expected, the coefficient estimates differed substantially. Furthermore, no consensus could be found in the literature regarding a procedure to back‐transform the coefficient estimates obtained from linear models performed on transformed datasets. This lack of consistency among coefficient estimates constitutes a major argument for model reformation over data transformation in biology.

show abstract

“…Pseudoreplication is a long recognized problem in ecology (Hurlbert ; Underwood ; Steel et al . ), in fisheries (Millar & Anderson ) and in aquaculture (Ling & Cotter ). This study confirms the presence of the issue also in aquaculture studies: in one‐third of the studies, blurred design descriptions hindered clear conclusions about the presence of pseudoreplication, but in the remaining 44 studies, 27% displayed pseudoreplication in some form.…”

Section: Discussionmentioning

confidence: 99%

Issues of aquatic experimentation in research on marine finfish larval production in Europe and North America

Nicolaisen

2017

Reviews in Aquaculture

View full text Add to dashboard Cite

The aim of this review was to examine practices in academic research focusing on production of larvae of European marine fish species. The literature spanned peer‐reviewed articles over the years 1977–2014 for the main commercially produced marine species: Atlantic cod, halibut, sea bream, sea bass and turbot. Search hits were categorized into the main themes basic biology, natural habitat studies and aquaculture, and the latter theme further subdivided into nutrition, health, reproduction and production factor studies. In total, 74 of the production factor studies were thoroughly examined in terms of overall research structure, study designs, methods, analyses and reporting practices. A picture of a fragmented whole with many isolated contributors and limited cooperation appeared, and despite the multivariate nature and high complexity of the area, isolated and low replicated studies focusing on one variable at the time dominated. Randomization and blinding procedures were rarely accounted for, and a need to standardize practical conduct of research in terms of physical infrastructure, biological material used, control of tank environment, sampling, tank tending and response measurements appeared. Issues of statistical analysis were also apparent, spanning the choice of main techniques and follow‐up contrast procedures and Pseudoreplication. Additionally, the reporting of designs, analyses and results was often incomplete. Consequences of the different issues are discussed, and possible remedies proposed. The findings are in accordance with conclusions from contemporary studies addressing scientific quality and methodological issues at other areas of research.

show abstract

Applied statistics in ecology: common pitfalls and simple solutions

Cited by 63 publications

References 42 publications

Joint species distribution modelling for spatio‐temporal occurrence and ordinal abundance data

Joint species distribution modelling for spatio‐temporal occurrence and ordinal abundance data

Count data in biology—Data transformation or model reformation?

Issues of aquatic experimentation in research on marine finfish larval production in Europe and North America

Contact Info

Product

Resources

About