Bringing Ecologists and Statisticians Together for the Conservation of Endangered Species

Post provided by Cecilia Pinto and Luigi Spezia

The Benefits of High Frequency Data

One of the tagged flapper skates showing the three different kinds of tags. ©Cecilia Pinto

One of the tagged flapper skates showing the three different kinds of tags. ©Cecilia Pinto

High frequency data, like those obtained from individual electronic tags, carries the potential of giving us detailed information on the behaviour of species at the individual level. Such data are particularly useful for marine species, as we can’t observe them directly for long periods of time.

Understanding how individuals use water columns – both at daily and seasonal scales – can help define conservation measures such as restricting fishing activity to reduce by-catch or defining protected areas to help recovering populations or protect spawning and nursery areas. High frequency data have become popular as they give insight to detailed individual foraging behaviour and therefore the specific energetic needs that are linked to reproduction and fitness. Continue reading

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

The Value of Information: Does More Data Mean Better Decisions?

Post provided by Dr Stefano Canessa

Applied ecology can be defined as scientific knowledge that helps in making good management decisions. Scientists have a natural desire to collect information, managers want that information so that they know they are doing the right thing, and both generally act under the assumption that more information equals better decisions. This is generally correct, since information helps us make, well, informed decisions. Therefore, when our ecological knowledge is uncertain (which is practically always the case) we usually advocate further research.

On the other hand, however, information comes at a cost. It may cost money to collect it and take time to set up studies: both are usually in short supply. We can’t learn everything and often the information we can actually collect is still imperfect. So how do we determine if that additional piece of information we’d like to have is really valuable for our management?

In ‘When do we need more data? A primer on calculating the value of information for applied ecologists’ , Stefano Canessa and colleagues provide a tutorial to the calculation of value of information (VOI) for applied ecologists and managers who would like to know more about it, but are not familiar with decision-theoretic principles and notation.

What is ‘Value of Information’?

In decision analysis, the value of information is the improvement in the outcomes of our actions that we would expect if we could reduce or eliminate uncertainty before making a decision. Previously applied in engineering, economics and healthcare planning, VOI is also intuitively appealing for environmental management, where decisions must be made in the face of ubiquitous uncertainty.  Knowing the value of information can assist in designing monitoring and experimental programs, implementing adaptive management and prioritising sources of uncertainty. In other words, it can help applied ecologists and conservation managers find a focused, transparent way to address the inevitable need for “more data”.

An increasing number of studies are applying VOI to conservation management; however, in spite of its potential the technique is still underused in real-world applications, particularly beyond the small community of applied ecologists trained in decision-analytic methods.

Click Image to begin a Prezi Presentation on Value of Information

Click Image to begin a Prezi Presentation on Value of Information

In summary, three things determine the value of information:

  1. How much we already know (the more we know, the less beneficial it is to collect more information)
  2. Whether and how we would react to that extra information by changing actions, and how much better would the updated action be
  3. How good is the information we can actually get (think about sample sizes, imperfect detection, time lags, etc)

Continue reading

Understanding and Presenting YOUR Data

A Beginner’s Guide to Data Exploration and Visualisation with R
by Elena N. Ieno and Alain F. Zuur

A Beginner's Guide to Data ExplorationIn 2010 Alain Zuur, Elena Ieno and Chris Elphick published a paper in Methods in Ecology and Evolution entitled ‘A protocol for data exploration to avoid common statistical problems‘ (Volume 1, Issue 1). Little did they know at the time that this paper would become one of the journal’s all-time top downloaded and top cited papers, with a total of 22,472 downloads between 2010 and 2014.

Based on this success they decided to extend the material in the paper into a book.

Zuur and his colleagues at Highland Statistics ltd. give about 25 five-day statistics courses per year. Their typical audience consists of biological scientists at the post-graduate and post-doctoral levels. Early on in each course they have the following conversation with the participants:

Speaker: “Do you review submitted manuscripts for journals?”
Audience: “Yes.”
Speaker: “Do you like the statistical part of these manuscripts?”
Audience: “No!”
Speaker: “Do you understand the statistical part?”
Audience: “Not always.”

What if there were ways you could make reviewing your paper easier and more enjoyable for reviewers? What if making your manuscript easier to understand and nicer to read would increase the likelihood of your work being published?

A Beginner’s Guide to Data Exploration and Visualisation with R explains how you can do exactly that! Alain Zuur and Elena Ieno use ecological datasets to discuss the data exploration and visualisation tools you can use to make your paper simpler for readers and reviewers to understand. The authors also explain how to visualise the results of statistical models, an important aspect of scientific papers. Continue reading