Dangers of forcing regressions through the origin

17 10 2017

correlationsI had an interesting ‘discussion’ on Twitter yesterday that convinced me the topic would make a useful post. The specific example has nothing whatsoever to do with conservation, but it serves as a valuable statistical lesson for all concerned about demonstrating adequate evidence before jumping to conclusions.

The data in question were used in a correlation between national gun ownership (guns per capita) and gun-related deaths and injuries (total deaths and injuries from guns per 100,000 people) (the third figure in the article). As you might intuitively expect, the author concluded that there was a positive correlation between gun-related deaths and injuries, and gun ownership:

image-20160307-30436-2rzo6k

__

Now, if you’re an empirical skeptic like me, there was something fishy about that fitted trend line. So, I replotted the data (available here) using Plot Digitizer (if you haven’t yet discovered this wonderful tool for lifting data out of figures, you would be wise to get it now), and ran a little analysis of my own in R:

Rplot01

Just doing a little 2-parameter linear model (y ~ α + βx) in R on these log-log data (which means, it’s assumed to be a power relationship), shows that there’s no relationship at all — the intercept is 1.3565 (± 0.3814) in log space (i.e., 101.3565 = 22.72), and there’s no evidence for a non-zero slope (in fact, the estimated slope is negative at -0.1411, but it has no support). See R code here.

Now, the author pointed out what appears to be a rather intuitive requirement for this analysis — you should not have a positive number of gun-related deaths/injuries if there are no guns in the population; in other words, the relationship should be forced to go through the origin (xy = 0, 0). You can easily do this in R by using the lm function and setting the relationship to y ~ 0 + x; see code here). Read the rest of this entry »





Cartoon guide to biodiversity loss XLIV

9 10 2017

Here’s another set of biodiversity cartoons to make you giggle, groan, and contemplate the Anthropocene extinction crisis. See full stock of previous ‘Cartoon guide to biodiversity loss’ compendia here.

Read the rest of this entry »





Four decades of fragmentation

27 09 2017

fragmented

I’ve recently read perhaps the most comprehensive treatise of forest fragmentation research ever compiled, and I personally view this rather readable and succinct review by Bill Laurance and colleagues as something every ecology and conservation student should read.

The ‘Biological Dynamics of Forest Fragments Project‘ (BDFFP) is unquestionably one of the most important landscape-scale experiments ever conceived and implemented, now having run 38 years since its inception in 1979. Indeed, it was way ahead of its time.

Experimental studies in ecology are comparatively rare, namely because it is difficult, expensive, and challenging in the extreme to manipulate entire ecosystems to test specific hypotheses relating to the response of biodiversity to environmental change. Thus, we ecologists tend to rely more on mensurative designs that use existing variation in the landscape (or over time) to infer mechanisms of community change. Of course, such experiments have to be large to be meaningful, which is one reason why the 1000 km2 BDFFP has been so successful as the gold standard for determining the effects of forest fragmentation on biodiversity.

And successful it has been. A quick search for ‘BDFFP’ in the Web of Knowledge database identifies > 40 peer-reviewed articles and a slew of books and book chapters arising from the project, some of which are highly cited classics in conservation ecology (e.g., doi:10.1046/j.1523-1739.2002.01025.x cited > 900 times; doi:10.1073/pnas.2336195100 cited > 200 times; doi:10.1016/j.biocon.2010.09.021 cited > 400 times; and doi:10.1111/j.1461-0248.2009.01294.x cited nearly 600 times). In fact, if we are to claim any ecological ‘laws’ at all, our understanding of fragmentation on biodiversity could be labelled as one of the few, thanks principally to the BDFFP. Read the rest of this entry »





A gender-diverse lab is a good lab

18 09 2017

sexism

Another little expurgated teaser from my upcoming book with Cambridge University Press.

My definition of a ‘lab’ is simply a group of people who do the science in question — and people are indeed a varied mob. I’d bet that most scientists do not necessarily give much thought to the diversity of the people in their lab, and instead probably focus more on obtaining the most qualified and cleverest people for the jobs that need doing. There are probably few of us who are overtly racist, sexist, or otherwise biased against or for certain types of people.

But the problem is not that scientists tend to exclude certain types of people deliberately based on negative stereotypes; rather, it concerns more the subconscious biases that might lurk within, and about which unfortunately most of us are blissfully unaware. But a scientist should be aware of, and seek to address, these hidden biases.

I acknowledge that as a man, I am stepping onto thin ice even to dare to discuss the thorny issue of gender inequality in science today, for it is a massive topic that many, far more qualified people are tackling. But being of the male flavour means that I have to, like an alcoholic, admit that I have a problem, and then take steps to resolve that problem.

Read the rest of this entry »





Less snow from climate change pushes evolution of browner birds

7 09 2017
© Bill Doherty

© Bill Doherty

Climate changes exert selective pressures on the reproduction and survival of species. A study of tawny owls from Finland finds that the proportion of two colour morphs varies in response to the gradual decline of snowfall occurring in the boreal region.

Someone born in the tropics who travels to the Antarctic or the Himalaya can, of course, stand the cold (with a little engineering help from clothing, however). The physiology of our body is flexible enough to tolerate temperatures alien to those of our home. We can acclimate and, if we are healthy, we can virtually reside anywhere in the world.

However, modern climate change is steadily altering the thermal conditions of the native habitats of many species. Like us, some can live up to as much heat or cold as their genetic heritage permits, because each species can express a range of morphological, physiological, and behavioural variation (plasticity). Others can modify their genetic make-up, giving way to novel species-specific features or genotypes (evolution).

When genetic changes are speedy, that is, within a few generations, we are witnessing ‘microevolution’ — in contrast to ‘macroevolution’ across geological time scales as originally reported by Darwin and Wallace (1). To date, the detection of microevolution in response to modern climate change remains elusive, and many studies claiming so seem to lack the appropriate data to differentiate microevolution from phenotypic plasticity (i.e., the capacity of a single genotype to exhibit variable phenotypes in different environments) (2, 3). Read the rest of this entry »





Which countries protect the most of their land?

1 09 2017

forestOne potentially useful metric to measure how different nations value their biodiversity is just how much of a country’s land its government sets aside to protect its natural heritage and resources. While this might not necessarily cover all the aspects of ‘environment’ we need to explore, we know from previous research that the more emphasis a country places on protecting its biodiversity, the more it actually achieves this goal. This might sound intuitive, but there is no shortage of what have become known as ‘paper parks’ around the world, which are essentially only protected in principle, but not in practice.

For example, if a national park or some other type of protected area is not respected by the locals (who might rightly or wrongly perceive them as a limitation of their ‘rights’ of exploitation), or is pilfered by corrupt government officials in cahoots with extractive industries like logging or mining, then the park does not do well in protecting the species it was designed to safeguard. So, even though the proportion of area protected within a country is not a perfect reflection of its environmental performance, it tends to indicate to what extent its government, and therefore, its people, are committed to saving its natural heritage.

Read the rest of this entry »





When to appeal a rejection

26 08 2017

BegA modified excerpt from my upcoming book for you to contemplate after your next rejection letter.

This is a delicate subject that requires some reflection. Early in my career, I believed the appeal process to be a waste of time. Having made one or two of them to no avail, and then having been on the receiving end of many appeals as a journal editor myself, I thought that it would be a rare occasion indeed when an appeal actually led to a reversal of the final decision.

It turns out that I was very wrong, but not in terms of simple functional probability that you might be thinking. Ironically, the harder it is to get a paper published in a journal, the higher the likelihood that an appeal following rejection will lead to a favourable outcome for the submitting authors. Let me explain. Read the rest of this entry »