There’s Madness in our Methods: Improving inference in ecology and evolution

Post provided by JARROD HADFIELD

Last week the Center for Open Science held a meeting with the aim of improving inference in ecology and evolution. The organisers (Tim Parker, Jessica Gurevitch & Shinichi Nakagawa) brought together the Editors-in-chief of many journals to try to build a consensus on how improvements could be made. I was brought in due to my interest in statistics and type I errors – be warned, my summary of the meeting is unlikely to be 100% objective.

True Positives and False Positives

The majority of findings in psychology and cancer biology cannot be replicated in repeat experiments. As evolutionary ecologists we might be tempted to dismiss this because psychology is often seen as a “soft science” that lacks rigour and cancer biologists are competitive and unscrupulous. Luckily, we as evolutionary biologists and ecologists have that perfect blend of intellect and integrity. This argument is wrong for an obvious reason and a not so obvious reason.

We tend to concentrate on significant findings, and with good reason: a true positive is usually more informative than a true negative. However, of all the published positives what fraction are true positives rather than false positives? The knee-jerk response to this question is 95%. However, the probability of a false positive (the significance threshold, alpha) is usually set to 0.05, and the probability of a true positive (the power, beta) in ecological studies is generally less than 0.5 for moderate sized effects. The probability that a published positive is true is therefore 0.5/(0.5+0.05) =91%. Not so bad. But, this assumes that the hypotheses and the null hypothesis are equally likely. If that were true, rejecting the null would give us very little information about the world (a single bit actually) and is unlikely to be published in a widely read journal. A hypothesis that had a plausibility of 1 in 25 prior to testing would, if true, be more informative, but then the true positive rate would be down to (1/25)*0.5/((1/25)*0.5+(24/25)*0.05) =29%. So we can see that high false positive rates aren’t always the result of sloppiness or misplaced ambition, but an inevitable consequence of doing interesting science with a rather lenient significance threshold. Continue reading

Issue 6.11

Issue 6.11 is now online!

The November issue of Methods is now online!

This month’s issue contains two Applications articles and one Open Access article, all of which are freely available.

mvMORPH: A package of multivariate phylogenetic comparative methods for the R statistical environment which allows fitting a range of multivariate evolutionary models under a maximum-likelihood criterion. Its use can be extended to any biological data set with one or multiple covarying continuous traits.

Low-cost soil CO2 efflux and point concentration sensing systems: The authors use commercially available, low-cost and low-power non-dispersive infrared (NDIR) CO2 sensors to develop a soil CO2 efflux system and a point CO2 concentration system. Their methods enable terrestrial ecologists to substantially improve the characterization of CO2 fluxes and concentrations in heterogeneous environments.

This month’s Open Access article comes from Jolyon Troscianko and Martin Stevens. In ‘Image calibration and analysis toolbox – a free software suite for objectively measuring reflectance, colour and pattern‘ they introduce a toolbox that can convert images to correspond to the visual system (cone-catch values) of a wide range of animals, enabling human and non-human visual systems to be modelled. The toolbox is freely available as an addition to the open source ImageJ software and will considerably enhance the appropriate use of digital cameras across multiple areas of biology. In particular, researchers aiming to quantify animal and plant visual signals will find this useful. This article received some media attention upon Early View publication over the summer. You can read the Press Release about it here.

Our November issue also features articles on Population Genetics, Macroevolution, Modelling species turnover, Abundance modelling, Measuring stress and much more. Continue reading

Issue 6.10

Issue 6.10 is now online!

The October issue of Methods is now online!

This month’s issue contains two Applications articles and one Open Access article, all of which are freely available.

letsR: A package for the R statistical computing environment, designed to handle and analyse macroecological data such as species’ geographic distributions and environmental variables. It also includes functions to obtain data on species’ habitat use, description year and current as well as temporal trends in conservation status.

Cleaning Oil from Seabirds: The authors assess the efficacy of sea water as an alternative to fresh water for cleaning oil from seabirds’ feathers. Results indicate that for oiled feathers, a sea water wash/rinse produced clean, low BAI/unclumped feathers with minimal particulate residue.

Stefano Canessa et al. provide this month’s only Open Access article. In ‘When do we need more data? A primer on calculating the value of information for applied ecologists‘ the authors guide readers through the calculation of Value of Information (VoI) using two case studies and illustrate the use of Bayesian updating to incorporate new information. Collecting information can require significant investments of resources; VoI analysis assists managers in deciding whether these investments are justified. The authors also wrote a blog post on VoI which you can find here.

Our October issue also features articles on Niche Modelling, Population Ecology, Spatial Ecology, Conservation, Monitoring and much more. Continue reading

Issue 6.9

Issue 6.9 is now online!

The September issue of Methods is now online!

This month’s issue contains one Applications article and two Open Access articles, all of which are freely available.

POPART: An integrated software package that provides a comprehensive implementation of haplotype network methods, phylogeographic visualisation tools and standard statistical tests, together with publication-ready figure production. The package also provides a platform for the implementation and distribution of new network-based methods.

Michalis Vardakis et al. provide this month’s first Open Access article. In ‘Discrete choice modelling of natal dispersal: ‘Choosing’ where to breed from a finite set of available areas‘ the authors show how the dispersal discrete choice model can be used for analysing natal dispersal data in patchy environments given that the natal and the breeding area of the disperser are observed. This model can be used for any species or system that uses some form of discrete breeding location or a certain degree of discretization can be applied.

Our September issue also features articles on Animal Movement, Population Dynamics, Statistical Ecology, Biodiversity, Conservation Biology and much more. Continue reading

Study Finds Black Bears in Yosemite Forage Primarily on Plants and Nuts

Below is a press release about the Methods paper ‘Measuring the realized niches of animals using stable isotopes: from rats to bears‘ taken from the University of California, San Diego:

©PLF73 (Click image to see original)

Animal proteins only make up a small part of a black bear’s diet. ©PLF73 (Click image to see original)

Black bears in Yosemite National Park that don’t seek out human foods subsist primarily on plants and nuts, according to a study conducted by biologists at UC San Diego who also found that ants and other sources of animal protein, such as mule deer, make up only a small fraction of the bears’ annual diet.

Their study, published in Methods in Ecology and Evolution, might surprise bear ecologists and conservationists who had long assumed that black bears in the Sierra Nevada rely on lots of protein from ants and other insects because their remains are frequently found in bear feces. Instead, the researchers believe that bears likely eat ants for nutrient balance. Continue reading

Issue 6.8

Issue 6.8 is now online!

The August issue of Methods is now online!

This month’s issue contains two Applications article and one Open Access article, all of which are freely available.

LEA: This R package enables users to run ecological association studies from the R command line. It can perform analyses of population structure and genome scans for adaptive alleles from large genomic data sets. The package derives advantages from R programming functionalities to adjust significance values for multiple testing issues and to visualize results.

PIPITS: An open-source stand-alone suite of software for automated processing of Illumina MiSeq sequences for fungal community analysis. PIPITS exploits a number of state of the art applications to process paired-end reads from quality filtering to producing OTU abundance tables.

Giovanni Strona and Joseph Veech provide this month’s Open Access article. Many studies have focused on nestedness, a pattern reflecting the tendency of network nodes to share interaction partners, as a method of measuring the structure of ecological networks. In ‘A new measure of ecological network structure based on node overlap and segregation‘ the authors introduce a new statistical procedure to measure both this kind of structure and the opposite one (i.e. species’ tendency against sharing interacting partners).

In addition to this, our August issue features articles on Estimating Diversity, Ecological Communities and Networks, Genetic Distances and Immunology. Continue reading

Impact of Flooding on Wetlands Measurable via Low-Cost Approach

Below is a press release about the Methods paper ‘Testing sea-level rise impacts in tidal wetlands: a novel in situ approach‘ taken from the University of Alabama:

IMG_0175 (2)

Researchers installed adjustable enclosures near the shoreline to simulate low, medium & high flooding levels. © Dr Eric Sparks, Mississippi State University

Scientists designed a new, on-site method for studying potential impacts rising sea levels can have on vital wetlands, said a University of Alabama researcher who led a study publishing in Methods in Ecology and Evolution today describing the modifiable apparatuses.

Primarily using materials available at the local hardware store, the scientists, including UA’s Dr Julia Cherry, designed, constructed and tested low-cost enclosures, called weirs, to realistically simulate three flooding levels on coastal wetlands. Simulating impacts of sea level rise on-site and at larger scales had previously proven difficult.

“I hope this provides other researchers with a template to ask their questions and to improve upon the method we’ve documented to do bigger and better coastal wetland studies,” said Cherry, an associate professor in UA’s New College and its biological sciences department. Continue reading

Issue 6.7

Issue 6.7 is now online!

The July issue of Methods is now online!

This month’s issue contains two Applications article and one Open Access article, all of which are freely available.

fuzzySim: Binary similarity indices are widely used in ecology. This study proposes fuzzy versions of the binary similarity indices most commonly used in ecology, so that they can be directly applied to continuous (fuzzy) rather than binary occurrence values, producing more realistic similarity assessments. fuzzySim is an open source software package which is also available for R. A freely accessible, web-based analysis tool for complex activity data, provides cloud-based and automatic computation of daily aggregates of various activity parameters based on recorded immersion data. It provides maps and graphs for data exploration, download of processed data for modelling and statistical analysis, and tools for sharing results with other users.

Anna Sturrock et al. provide this month’s Open Access article. In ‘Quantifying physiological influences on otolith microchemistry‘ the authors test relationships between otolith chemistry and environmental and physiological variables. The influence of physiological factors on otolith composition was particularly evident in Sr/Ca ratios, the most widely used elemental marker in applied otolith microchemistry studies. This paper was reported on in the media recently. You can read more about it here.

Our July issue also features articles on Monitoring, Remote Sensing, Conservation, Genetics and three papers on Statistics. Continue reading

Issue 6.6

Issue 6.6 is now online!

The June issue of Methods is now online!

This month’s issue contains one Applications article and one Open Access article.

VirtualCom: A simple and readily usable tool that will help to resolve theoretical and methodological issues in community ecology. VirtualCom simulates the evolution of the pool of regionally occurring species, the process-based assembly of native communities and the invasion of novel species into native communities. One of the authors of this Application is the 2014 Robert May Young Investigator Prize Winner, Laure Gallien.

Calibrating animal-borne proximity loggers, this month’s only Open Access article, comes from Christian Rutz et al. The authors calibrated a recently developed digital proximity-logging system (‘Encounternet’) for deployment on a wild population of New Caledonian crows. They show that, using signal-strength information only, it is possible to assign crow encounters reliably to predefined distance classes, enabling powerful analyses of social dynamics. Their study demonstrates that well-calibrated proximity-logging systems can be used to chart social associations of free-ranging animals over a range of biologically meaningful distances.

Our June issue also features articles on Phylogenetic MethodsPhysiological Ecology, Biomonitoring and Conservation, Species Distribution Monitoring and Bioinformatics. Continue reading

Issue 6.5

Issue 6.5 is now online!

The May issue of Methods is now online!

We have two freely available articles this month: one Application and one Open Access Article.

rSPACE: An open-source R package for implementing a spatially based power analysis for designing monitoring programs. This method incorporates information on species biology and habitat to parameterize a spatially explicit population simulation.

Tim Lucas et al. provide this month’s Open Access article: A generalised random encounter model for estimating animal density with remote sensor data. The authors have developed a Generalised Random Encounter Model (gREM) to estimate absolute animal density from count data from both camera traps and acoustic detectors. They show that gREM produces accurate estimates of absolute animal density for all combinations of sensor detection widths and animal signal widths. This model is applicable for count data obtained in both marine and terrestrial environments, visually or acoustically. It could be used for big cats, sharks, birds, echolocating bats, cetaceans and much more. Continue reading