Skip to main content

TR Memescape

  • Talk Rational: Seriously, we call it that.

Recent Posts

11
Don't even try to show him distribution normalization, Pingu, lol.

Dave might be tending to use "normal" colloquially, as in "usual", which wouldn't be surprising. What is also not surprising for a person who claims to be so good at science, who misunderstands experimental controls and hypothesis testing, to not have a clue what the normal distribution vertical axis variable might even refer to.

Good luck.

Also, the distribution is where the "normal" comes from so from a colloquial perspective, his statement was fair even if it demonstrates ignorance of the deeper concepts.
12
Computers and Technology / Re: Digital camera?
Last post by Testy Calibrate -
Lenses are insane. But cool about the camera.
13
Don't even try to show him distribution normalization, Pingu, lol.

Dave might be tending to use "normal" colloquially, as in "usual", which wouldn't be surprising.

Yes.  But then his definition wouldn't be true.  You wouldn't get much of a bell curve if you only included people with the most "usual" measurements.

What is also not surprising for a person who claims to be so good at science, who misunderstands experimental controls and hypothesis testing, to not have a clue what the normal distribution vertical axis variable might even refer to.

Good luck.

He has never understood variability.  And why it's important if you want to generalise.  For instance, if, instead of just trying to figure out how to live well on your own property, which is entirely laudable, you are trying to Save Agriculture And Thereby Save The World.
14
Politics and Current Events / Re: Trumpocalypse
Last post by meepmeep -


Most of us would "rather not," but ignoring this crazy piece of shit who happens to be President of the goddamn United States doesn't make it all go away, you fucking coward.
15
Don't even try to show him distribution normalization, Pingu, lol.

Dave might be tending to use "normal" colloquially, as in "usual", which wouldn't be surprising. What is also not surprising for a person who claims to be so good at science, who misunderstands experimental controls and hypothesis testing, to not have a clue what the normal distribution vertical axis variable might even refer to.

Good luck.
16
So I guess the age of the people we are measuring matters

Lol

Well, it matters that we don't just include "normal" measurements, which is what you said. 

And yes, if we plotted the age of children, you would probably not get a bell-curve, unless they were all the same age, or you selected your children by some process that gave you a normal distribution of ages.

Why don't you actually read my post?

I guess you'd rather scoff than try to engage with it, right?
17
i'm not going to play 20 questions with you because it is obvious that my answers will never be good enough for you.  if you have a point, just make it.

I already did.

But as I predicted, you didn't read it.
18
On a side note from Pingu's point: the reason why "It's because most things have lots of causes - some making them bigger;, better; stronger; safer; some making them smaller; worse; weaker; more dangerous." tends to form a bell curve rather than, say, a triangle curve (which you'd see, for example, when rolling a pair of dice), is called the Central Limit Theorem, a strengthening of the Law of large numbers.

To be fully cautious: that by itself isn't really enough to show that heights should form a bell curve. You'd need something along the lines of "These factors are independent, and there are enough large factors of approximately equal size" to get there.

Right.  And it's an important reason why the tails of distributions are sometimes quite lumpy.  Positive feedback loops often kick in when the causal factors themselves are at extremes.

But with enough factors, even non-independent factors can result in a bell curve - negative feedback loops can often occur as often as positive ones.

And sometimes we can even "unmix" the factors, by plotting everything against everything else, and jiggling the matrix so that the axis are maximally non-gaussian, and you can even extract the underlying binary distributions!

But I'm sure you can explain that much better, uncool!

And of course many things do NOT have a normal distribution.  Crap dishwashers have a Poisson distribution without a low tail. 
19
 i'm not going to play 20 questions with you because it is obvious that my answers will never be good enough for you.  if you have a point, just make it.
20
Computers and Technology / Re: Digital camera?
Last post by Worldtraveller -
Welp, pulled the trigger on a Nikon D3300. Looks like a pretty good deal, via Ebay.

Of course, the lens I'm eyeballing costs ~3x as much as the initial camera and 2 lenses that come with it. :D