Page 1 of 1

Urban pigeons lose toes due to human activity

Posted: Wed Nov 13, 2019 11:34 pm
by Fishnut
New research finds that foot mutilations in urban pigeons are associated with increased human population density, and noise and air pollution. The abstract says that there's a correlation in mutilations with abundance of hairdressers but,
After excluding correlated predictors, we ran generalized linear models and applied a stepwise selection procedure of variables based on variance inflexion factors (see Material and methods for details). We identified noise pollution and air pollution as the main predictors of mutilation occurrence (Table1). No other predictors were significant, though the quadratic term of hairdresser density was close to significance (P=0.075, Fig. 2c). [my emphasis, p4]
I'd always thought you couldn't say that something was "close to significant" like this. And the results mean it is insignificant so to be highlighting hairdressers regardless has me a little concerned.

The authors also point out that,
there are also more hairdressers where human population density is higher (r=0.57 see Fig.S2 in Supplementary material), leading to a potential indirect effect of human population density [p4]
It's important research and I like the point that "There is no ethical reason for accepting that pigeons should suffer from mutilation due to human development without trying to reduce their pain" though I could have done without the highlighting of hairdressers without more robust evidence.

P.S. The paper is behind a paywall but if anyone wants a copy DM me.

Re: Urban pigeons lose toes due to human activity

Posted: Thu Nov 14, 2019 12:08 am
by dyqik
The p<=0.05 equals significance thing is just an arbitrary threshold, so p<=0.075 isn't much worse than that.

We often use 5 sigma for detection of astronomical sources, which is p<=0.0000003. Any higher than that isn't deemed significant.

Re: Urban pigeons lose toes due to human activity

Posted: Thu Nov 14, 2019 1:03 am
by Fishnut
I know that it's an arbitrary cut-off but once that cutoff has been determined it's not, well, good form, to say "ah, it's close enough" and start interpreting the meaning of the result. In biology 0.05 is a standard cut-off but it is a cut-off and I seem to remember watching some videos by a stats guy (wish I could remember his name, it was part of a Coursera course I think - just remembered, it's Daniel Lakens) who explained that for reasons I can't remember values around the 0.05 cut-off are more likely to be not significant than they are to be significant.

And even if that isn't the case, if you've decided that everything above 0.05 is not significant you can't really turn around and go "but this result would be really cool if it was significant so let's discuss it anyway."

Re: Urban pigeons lose toes due to human activity

Posted: Thu Nov 14, 2019 7:25 am
by basementer
Well.
In this model, which is predicting something like toe injuries per pigeon, the coefficient of the quadratic term of hairdresser density would presumably have to have nonsensical units like
((toe injuries) * m^4) / (hairdressers^2)*pigeons)
But I can't see the paper, so I might well be misrepresenting the authors.

Re: Urban pigeons lose toes due to human activity

Posted: Thu Nov 14, 2019 7:27 am
by Woodchopper
Density of hairdressers is an odd variable.

It seems like a reasonable proxy for population density. Almost everyone gets a haircut sooner or later.

However I’m surprised that a proxy would be needed in somewhere like Paris. In my experience French demographic statistics are very comprehensive. They are the nation that invented sociology. So the authors should have been able to obtain accurate statistics on actual population density.

I can’t access the paper though, so perhaps they explain why they used that variable.

Re: Urban pigeons lose toes due to human activity

Posted: Thu Nov 14, 2019 8:43 am
by Turdly
I've had a quick look at the paper.

The "close to significant" thing is a bit naughty. As stated up thread, the choice of confidence level is arbitrary but once you have chosen it you have to stick to it. I've taken students to task in the past for writing "treatment x (p=0.051) was not significant" at one point then "treatment x (p=0.052) was close to significant" later because it fit their narrative.

The hairdresser thing is a bit odd. From the intro it appears to have considered as there is evidence that hair can be a cause of injury.
However, a close observation of foot and toe injuries systematically reveals the presence or former presence of a string or more generally human hair caught on digits (Skandrani et al., 2018), a problem called ‘stringfeet’. The string or hair tightens more and more and eventually cut off the blood supply, further necrosis causing the toe to fall off.
However, the authors note that population density and hairdresser numbers are also reasonably correlated.

Re: Urban pigeons lose toes due to human activity

Posted: Thu Nov 14, 2019 9:18 am
by cvb
Turdly wrote:
Thu Nov 14, 2019 8:43 am
I've had a quick look at the paper.

The "close to significant" thing is a bit naughty. As stated up thread, the choice of confidence level is arbitrary but once you have chosen it you have to stick to it. I've taken students to task in the past for writing "treatment x (p=0.051) was not significant" at one point then "treatment x (p=0.052) was close to significant" later because it fit their narrative.

The hairdresser thing is a bit odd. From the intro it appears to have considered as there is evidence that hair can be a cause of injury.
However, a close observation of foot and toe injuries systematically reveals the presence or former presence of a string or more generally human hair caught on digits (Skandrani et al., 2018), a problem called ‘stringfeet’. The string or hair tightens more and more and eventually cut off the blood supply, further necrosis causing the toe to fall off.
However, the authors note that population density and hairdresser numbers are also reasonably correlated.
Okay that makes more sense. I was wondering what hairdressers had against pigeons.

Re: Urban pigeons lose toes due to human activity

Posted: Thu Nov 14, 2019 9:30 am
by tom p
Perhaps a nearly significant percentage of them throw old scissors away carelessly.

Re: Urban pigeons lose toes due to human activity

Posted: Thu Nov 14, 2019 12:20 pm
by Woodchopper
cvb wrote:
Thu Nov 14, 2019 9:18 am
Turdly wrote:
Thu Nov 14, 2019 8:43 am
I've had a quick look at the paper.

The "close to significant" thing is a bit naughty. As stated up thread, the choice of confidence level is arbitrary but once you have chosen it you have to stick to it. I've taken students to task in the past for writing "treatment x (p=0.051) was not significant" at one point then "treatment x (p=0.052) was close to significant" later because it fit their narrative.

The hairdresser thing is a bit odd. From the intro it appears to have considered as there is evidence that hair can be a cause of injury.
However, a close observation of foot and toe injuries systematically reveals the presence or former presence of a string or more generally human hair caught on digits (Skandrani et al., 2018), a problem called ‘stringfeet’. The string or hair tightens more and more and eventually cut off the blood supply, further necrosis causing the toe to fall off.
However, the authors note that population density and hairdresser numbers are also reasonably correlated.
Okay that makes more sense. I was wondering what hairdressers had against pigeons.
Yes, indeed.

Re: Urban pigeons lose toes due to human activity

Posted: Thu Nov 14, 2019 12:27 pm
by Martin Y
Having first idly wondered if pigeons like green jelly beans, I moved on to wonder if they like to line their nests with hair. But then I considered that the streets around hairdressers shops are not distinguished by telltale tufts of clipped hair and decided meh.

Re: Urban pigeons lose toes due to human activity

Posted: Thu Nov 14, 2019 4:06 pm
by Bird on a Fire
Pigeons tend not to line their nests with anything, really. They just build a flimsy pile of sticks on a ledge somewhere.

I have heard of chicks of other species getting tangled in hair, if they hatched in a nest lined with (usually horse) hair, and being unable to fly off so they starve to death. Pretty unusual though.

My wife and I both have long hair, and it certainly does manage to get wrapped around absolutely everything. I've occasionally wondered what caused pigeons to lose toes - getting snared might make sense.

Re: Urban pigeons lose toes due to human activity

Posted: Thu Nov 14, 2019 5:08 pm
by Boustrophedon
Bird on a Fire wrote:
Thu Nov 14, 2019 4:06 pm
Pigeons tend not to line their nests with anything, really. They just build a flimsy pile of sticks on a ledge somewhere.

I have heard of chicks of other species getting tangled in hair, if they hatched in a nest lined with (usually horse) hair, and being unable to fly off so they starve to death. Pretty unusual though.

My wife and I both have long hair, and it certainly does manage to get wrapped around absolutely everything. I've occasionally wondered what caused pigeons to lose toes - getting snared might make sense.
I would have thought that sitting on a fire would cause birds to lose toes too.

Re: Urban pigeons lose toes due to human activity

Posted: Thu Nov 14, 2019 7:46 pm
by Bird on a Fire
People think I have long hair and a beard because I'm some kind of hippy, but really it's just self-preservation.

Re: Urban pigeons lose toes due to human activity

Posted: Fri Nov 15, 2019 12:09 pm
by Gfamily