Storks Deliver Babies (p= 0.008) (2000) [pdf] | Hacker News Reader

Storks Deliver Babies (p= 0.008) (2000) [pdf]

web.archive.org

132 pointsWayToDoor61 comments

Open articleView on HN

Related: Lack of pirates is causing global warming:

https://www.forbes.com/sites/erikaandersen/2012/03/23/true-f...

Hah! I love this kind of stuff

REPORT: Average human has less than two arms

well, sure, after sharkweek!

There are no humans with three or more arms. Most humans have two arms. A few humans have one or zero arms. Thus, the mean/average number of arms should be between 1.9 and 2.0 but surely less then two.

(The median number of arms is two though)

Prove it.

first, I am aware of the stats so the lecture was a waste.

second, the posters name was sharkweek, if you went in the water during sharkweek you probably wouldn't have two arms, this is what is known as a joke based on an eponysterical name.

third, you are also incorrect in your statement that there are no humans with 3 arms https://www.google.com/search?q=baby+born+with+three+arms

You are right. I was wrong. You win this debate.

There are no humans with three or more arms

Never say never:

https://en.wikipedia.org/wiki/Polymelia

> Never say never:

Especially on the internet

Average human has more than 1 skeleton!

(pregnancy)

Some people own skeletons.

All own at least one. Some, two.

The extra ones are usually kept in the closet, I believe.

The average human has approximately one testicle and one boob.

You could also say that the most humans have an above-average number of arms

Also related: Spurious Correlations[0]. These are great and really drives home the idea that correlation neither equals or implies causation. At least one of these graphs implies that if the US wanted to reduce the number of drivers getting hit by trains, they simply just need to stop importing oil from Norway.

[0]: https://www.tylervigen.com/spurious-correlations

There are some cases where lack of correlation is actually evidence of causation.

As in the causal effect happens to cancel out a confounder, or something more fun? Do you have an example?

I was thinking of an air conditioner - a perfect A/C would mean the temperature is flat. Although hmm, I think R^2 is undefined rather than 0 there.

But basically, it's a "if you do it right, it'll look like you've done nothing at all" situation.

What variables are you correlating in that example? It seems to correlate just fine if your measuring energy inputed to heat the room vs work done by air conditioner.

This is rather silly (and not in the way intended).

The goal of the paper is to demonstrate a strong linear correlation which arises without a trivial confounder (such as age influencing both reading age and shoe size in children). However, the confounder in the case of storks (country land area influences number of storks and absolute number of births) IS trivial.

> One candidate for a potential confounding variable is land area: readers are invited to investigate this possibility using the data in table 1.

Yes - to be clear, the author knows this is the confounder. That's why it was silly of the author to claim that the paper would describe a correlation without a trivial confounder.

Land area isn't well-correlated with the other three columns in the table.

Yes it is! I calculated R = 0.58, 0.81, 0.92, respectively. These will be highly significant, comparable to the p-value in the headline.

https://docs.google.com/spreadsheets/d/1QTEB7KF1PXBV_TjgBsuV...

My mom was an OB/GYN nurse for several decades. She worked in Active Labor for many years with a nurse named Alice Stork.

This Stork did indeed deliver thousands of babies.

An acquaintance of mine was a team manager for a government-welfare phone-support entity. The other two managers' names were D'eath and Bludd. My friend married a man named Paine.

So the phone-support teams were managed by Blood, Death, and Pain. It's one of the most wonderful matchups I've ever seen.

Sounds like a case of nominative determinism!

https://en.wikipedia.org/wiki/Nominative_determinism

Is this concept often mentioned in mainstream media? This is mentioned a lot around here. My hypothesis is that a "hn-bias" makes commenters more likely to mention it because they learned here that there is an explicit concept to arrange these funny coincidences, and moreover there is some quantitative work that has been done on the subject. But it could be a variant of frequency illusion/Baader-Meinhof phenomenon (what would be its name?).

Related: "The ASA Statement on p-Values: Context, Process, and Purpose"

- https://www.tandfonline.com/doi/full/10.1080/00031305.2016.1...

- https://news.ycombinator.com/item?id=30324223

Reminded me of this: https://tylervigen.com/spurious-correlations

I wouldn’t trust this article. How can only be 1 pair of storks in Belgium, while 5000 pairs in Hungary? I think the article needs better peer review before I accept the fact that storks deliver babies.

You're right, the only thing we can logically conclude from this study is that the number of storks known to humans correlates with birth rates, leading to the obvious conclusion that countries which are better at finding storks are also better at having procreative sex, probably due to the close similarity between the two behaviors.

Being a bird watcher is a prophylactic?

I'm happy to inform you that nowadays the storks in Belgium are sort of returning. There's two successful breeding programs happening: one in "Het Zwin" and one in "Planckendael". It is sadly true that the stork used to be quasi-extinct over here. Outside of these breeding programmes I've never seen one around.

Wow, that's crazy. I lived in Hungary and I saw them in every village all my life, they are always so cute, I'm glad they are back, hopefully there will be more babies born as well :)

Also, the argument that the Covid vaccines are safe and free of tracker chips is completely lacking.

At least we can all agree that it doesn't convincingly argue that horse dewormer is dangerous.

Calling a medicine that got the Nobel Prize for helping cure parasite dseases in Africa "a horse dewormer" is kinda racist.

I only see the first page when I click the article link. I found the full text online here: https://www.researchgate.net/profile/Robert-Matthews-2/publi...

Something wrong with your reader? It loaded normally for me.

It didn’t work for me either. Safari on iOS.

Thank you. I had the same problem on Chrome on ios.

Interesting. Must be a WebKit issue - I’m running safari on iOS

I cannot view it directly in firefox.

A direct view link in browser: https://docmadeeasy.com/v/266959716

Past related thread:

Storks Deliver Babies (p= 0.008) - https://news.ycombinator.com/item?id=24700606 - Oct 2020 (58 comments)

A hypothesis test allows us to say we rejected the null hypothesis.

It doesn't prove anything else besides that.

Another good resource on how (not) to mess up statistics is https://www.statisticsdonewrong.com/

I doubt storks care about national borders and this should probably be repeated using birth rates in general areas around stork populations ;)

Is it just me or is this article unreadable due to the atrocious typesetting? All the letters overlap.

My Firefox built in pdf viewer does not overlap the letters. Download the pdf and try another viewer.

The fusion of hydrogen into helium E = 0.008 mc ^2. Coincidence? I think not.

The confounding variable is how developed a country is.

It negatively influences both stork pairs (storks in Europe mostly live in old-style individual farming countryside) and birth rates.

The confounding variable is how large a country is (both size and population - two variables that are themselves related).

The papers “birth rate” is absolute: thousands of births per year in that country. It should be number of births per woman of childbearing years.

Likewise, the number of stork pairs is the absolute number in the country. It should be absolute number per square kilometer, or per square kilometer of stork habitat.

A geographically larger country will (generally) have more people and more storks.

Edit: there are also some outliers heavily influencing things: Poland and Turkey have tons of storks. Belgium, Denmark, Holland, and Switzerland don’t seem to have native storks.

> Poland and Turkey have tons of storks. Belgium, Denmark, Holland, and Switzerland don’t seem to have native storks.

This is the development factor I wrote about :) Storks in Europe live in symbiosis with individual farmers. As countries switch from millions of small farmers to thousands of massive industrialized farms - stork populations dwindle.

This is how it looks like: https://pl.wikipedia.org/wiki/Stary_Brus#/media/Plik:Stary_B...

Almost every house has its stork nest, people provide the scaffolding for the nest (usually on the house roof or on powerline pylons). Storks eat small rodents, insects and other pests (for example whenever farmers are plowing storks go behind the tractor and eat larva that are uncovered).

That corporate farms fail to provide stork nestage is a fail.

One corporate farm would have to provide hundreds nests to offset the individual farmers they replaced.

OK. So, provide. Mandate providing.

Storks in Denmark have dwindled to 1-2 pairs from a larger population.

This is of course also projected in the birth rate: https://www.macrotrends.net/countries/DNK/denmark/birth-rate

The human birth rate also declines the more developed a country is (per person, at least)

That's the point?

Ah I see, must have misread the first time round.

Reply on news.ycombinator.com