One of the recent posts:

"A study claiming a tenfold decrease in bugs splattered on evolutionary biologist Anders Møller’s windshield over two decades has been retracted."

NopIdoN 4 hours ago | parent | next [-]

What's a "tenfold decrease"?

	▲	dcrazy 4 hours ago \| parent \| next [-]
		Inverse of a tenfold increase, so 1/10 aka 0.1x previous number?
	▲	LPisGood 4 hours ago \| parent \| prev \| next [-]
		x -> x/10
	▲	jagged-chisel 3 hours ago \| parent \| prev [-]
		“10x less”

▲

raverbashing 5 hours ago | parent | prev [-]

Yeah this seems like a fundamentally anecdotal evidence, if bugs are splatting less on his car windshield

▲

bluGill 5 hours ago | parent [-]

Worse than anecdotal - even if there is real measured data: aerodynamics of windshields will have changed and have an effect and so we still cannot draw conclusions from this. Only if the experiment is more controlled (that is the same car driving on the same roads at the same speeds at the same time) could we draw a conclusion.

▲

jgraham 4 hours ago | parent | next [-]

We can't draw conclusions from that study because it's been retracted on the basis that data has been faked.

On the other hand there are other similar studies that reach similar conclusions, and specifically try to control for aerodynamics e.g. [1] which says

> The weak positive relationship between vehicle registration year and splat rate suggests that newer vehicles are more efficient at sampling insects than older vehicles.

i.e. they saw more insects on newer cars compared to older ones in the same time period.

In general ecology studies aren't like lab physics, you can't control every possible confounding variable; the systems are too complicated and studies ex-situ have their own limitations. But refusing to engage with the data we do have because it's not perfect isn't going to help you make better decisions, and doesn't represent some moral high ground.

[1] https://cdn.buglife.org.uk/2022/05/Bugs-Matter-2021-National...

	▲	bluGill 4 hours ago \| parent [-]
		I didn't mean we shouldn't engage with data at all. However there are so many possible confounding factors in this type of measurement that we should "take it with a lot of salt."

▲

dmoy 4 hours ago | parent | prev | next [-]

Same roads doesn't even control. If you lived in a town that e.g. changed the very local environment (say drained one specific swamp), the nearby roads my have less bugs for a very uninteresting reason

▲

shagie 4 hours ago | parent [-]

That is the interesting reason.

While the retraction brings into question the anecdotal evidence for the windscreen phenomenon ( https://en.wikipedia.org/wiki/Windshield_phenomenon ), there are other studies with other sampling approaches that support the global insect population collapse ( https://en.wikipedia.org/wiki/Decline_in_insect_populations ).

	▲	bluGill 3 hours ago \| parent [-]
		Even if this is true, we should still reject any falsified study that supports it.

▲

ErroneousBosh 3 hours ago | parent | prev [-]

I drive a 30 year old Range Rover and I can confirm that I have just about as many bug splats now as I did driving my equally un-aerodynamic Volvo 20 years ago.

Although the funniest one was driving through a cloud of moths on the A9 one summer about 30 years ago in my little Nissan, which hoovered up enough of them to choke the air filter and die on the next (fairly long and steep) hill. They were hell to get off the windscreen too.