0715: "Numbers"

This forum is for the individual discussion thread that goes with each new comic.

Moderators: Moderators General, Prelates, Magistrates

yet another steven
Posts: 47
Joined: Wed Jan 13, 2010 10:22 am UTC

Re: Numbers

Postby yet another steven » Wed Mar 17, 2010 11:51 am UTC

marcel_proust wrote:Cup size is a number? Who knew?

Sure: A=65, B=66, ...

yet another steven
Posts: 47
Joined: Wed Jan 13, 2010 10:22 am UTC

Re: Numbers

Postby yet another steven » Wed Mar 17, 2010 11:57 am UTC

jonamous wrote:There are probably a lot more than 200,000 with an IQ over 140.

If you take mean=100 and standard deviation=15, as is the most common calibration, then you're right: the probability that a random person's IQ is >= 140 is 1-Phi(40/15), where Phi is the standard normal cdf. Thus, one in 261 people is cleverer than that.

However, if the standard deviation of the test you're considering is 10, that figure becomes one in 31574 people. So for such a test it would be true that there are only about 200,000 people who score that high in the world.
Last edited by yet another steven on Wed Mar 17, 2010 12:06 pm UTC, edited 2 times in total.

SolkaTruesilver
Posts: 105
Joined: Mon Aug 20, 2007 11:30 am UTC

Re: Numbers

Postby SolkaTruesilver » Wed Mar 17, 2010 11:58 am UTC

Get out of my pants, Randal!!

User avatar
shoelessschuler
Posts: 33
Joined: Sun Jan 24, 2010 7:28 pm UTC
Location: Ontario
Contact:

Re: Numbers

Postby shoelessschuler » Wed Mar 17, 2010 12:27 pm UTC

Do I seriously have to drink 9 glasses of water a day? :(
"But my heart it don't beat, it don't beat the way it used to, and my eyes they don't see you no more, and my lips, they don't kiss, they don't kiss the way they used to and my eyes don't recognize you no more"

xepher
Posts: 83
Joined: Tue Mar 09, 2010 1:42 am UTC

Re: Numbers

Postby xepher » Wed Mar 17, 2010 12:33 pm UTC

shoelessschuler wrote:Do I seriously have to drink 9 glasses of water a day? :(

I remember hearing somewhere that about either 33% or 60% comes from food you eat. Well, it depends on the diet. Hence the drastic change in which one it was.

So... not as much as 9, considering the diet.

jonamous
Posts: 3
Joined: Wed Mar 17, 2010 11:39 am UTC

Re: Numbers

Postby jonamous » Wed Mar 17, 2010 12:35 pm UTC

yet another steven wrote:
jonamous wrote:There are probably a lot more than 200,000 with an IQ over 140.

If you take mean=100 and standard deviation=15, as is the most common calibration, then you're right: the probability that a random person's IQ is >= 140 is 1-Phi(40/15), where Phi is the standard normal cdf. Thus, one in 261 people is cleverer than that.

However, if the standard deviation of the test you're considering is 10, that figure becomes one in 31574 people. So for such a test it would be true that there are only about 200,000 people who score that high in the world.


You can't really consider the standard deviation to be 10. Triple Nine Society takes folks who score at or above the 99.9th percentile, which they estimate to be >149 IQ, assuming a standard deviation of 16 (which seems to be a common number when it comes to these tests).

Edit: Just read the first part of your post. ;) Sorry, had tunnel vision on the SD=10 part.

User avatar
Adacore
Posts: 2755
Joined: Fri Feb 20, 2009 12:35 pm UTC
Location: 한국 창원

Re: Numbers

Postby Adacore » Wed Mar 17, 2010 12:50 pm UTC

I had an officially administered IQ test when I was about 7 or 8 (I think my school was trying to get extra funding for having a 'gifted' student with the attached special needs budget) that gave me 147 - is this just a really common result for some reason? Perhaps test scoring methods are poorly designed - do they give poor resolution at high and low ranges, or something?
Last edited by Adacore on Wed Mar 17, 2010 12:54 pm UTC, edited 1 time in total.

ruthheart
Posts: 1
Joined: Wed Mar 17, 2010 12:30 pm UTC

Re: Numbers

Postby ruthheart » Wed Mar 17, 2010 12:54 pm UTC

The thing that bothers me about this comic is that the bra cup size graph stops at a "G", which makes it appear bras only go up to this cup size. Bra cups, in fact go up to at least a "M" cup. Most brick and mortar stores do not carry these larger sizes, as the American manufacturers do not produce them, but you can buy them online from British manufacturers (apparently women there are more likely to wear their correct size). I just hate anything that contributes to women wearing an incorrect bra size, and ignorance of larger sizes is a big factor in that.
Ruth

joshwa
Posts: 2
Joined: Wed Jan 06, 2010 11:30 pm UTC

Re: Numbers

Postby joshwa » Wed Mar 17, 2010 1:04 pm UTC

I posted this back in the Dangers thread:

http://github.com/joshwand/misc/blob/ma ... ildcard.rb

This will return a list of counts for any google wildcard search. Too lazy right now to hook it up to gnuplot, but it should be trivial.

elevul
Posts: 7
Joined: Sun Nov 16, 2008 4:33 pm UTC

Re: Numbers

Postby elevul » Wed Mar 17, 2010 1:05 pm UTC

katiekat wrote:The bra cup thing is actually really easy to explain. Among women, there are two main breast-size related communities on the internet where one may profess their size (and not for bragging):

1) A community of women who are smaller sized (ie, IBTC), who come together for support and tips on finding bras that fit and flatter and "give you curves"
Example: http://32aabra.com/

2) A community of women who are exceptionally busty, and come together for support and tips on finding bras that fit and are cute. (Just try finding well-fitting, cute bras when you're an E cup.)
Example: http://www.ugcs.caltech.edu/~cat/bra.html

Women who have "C" cups, which is generally indeed considered "the perfect size" need support neither in feeling better about themselves nor in finding bras or clothes that fit them. Thus they don't spend time on the internet talking about their breasts.

Nice explanation. ;)

yet another steven
Posts: 47
Joined: Wed Jan 13, 2010 10:22 am UTC

Re: Numbers

Postby yet another steven » Wed Mar 17, 2010 1:07 pm UTC

Adacore wrote:I had an officially administered IQ test when I was about 7 or 8 (I think my school was trying to get extra funding for having a 'gifted' student with the attached special needs budget) that gave me 147 - is this just a really common result for some reason?


No, that's actually pretty damn high :D Provided of course that they used a reliable test and took it properly, which I'll assume they did. Tests are usually less accurate for very high and very low scores; for example, if some alien would come up and say her (its?) IQ is 300, there would be no way to verify if that claim is correct.

EDIT: ah, I didn't see you referred to the spike in the graphs. It's a bit weird to speak of a score that less than one in thousand children reach as "a really common result" though :D

elevul
Posts: 7
Joined: Sun Nov 16, 2008 4:33 pm UTC

Re: Numbers

Postby elevul » Wed Mar 17, 2010 1:21 pm UTC

mirni wrote:2,830,000 for "i have too much time" ^^

-m-

3,760,000 for "I don't have enough time". :D

helo darqness
Posts: 52
Joined: Mon Jan 04, 2010 5:46 pm UTC

Re: Numbers

Postby helo darqness » Wed Mar 17, 2010 1:26 pm UTC

Its really bugging me that this thread isnt titled "Numbers" Discussion.

Jesus. For consistency's sake, change it!

User avatar
Arancaytar
Posts: 1642
Joined: Thu Mar 15, 2007 12:54 am UTC
Location: 52.44°N, 13.55°E
Contact:

Re: Numbers

Postby Arancaytar » Wed Mar 17, 2010 1:30 pm UTC

feedme wrote:haha I'm a big fan of these graphs he's done lately.

Also, the average internet user has a high IQ and a large penis? At least we're all truthful.


"I am completely honest" : 813,000
"I am honest": 3,850,000
"I am a liar": 7,060,000

I'm wondering what the real average is. It seems obvious that there is an IQ selection bias among the technophile population of the internet - but how much of it is due to the equally obvious liar selection bias? :P
Last edited by Arancaytar on Wed Mar 17, 2010 1:32 pm UTC, edited 1 time in total.
"You cannot dual-wield the sharks. One is enough." -Our DM.
Image

User avatar
littlelj
Posts: 140
Joined: Wed Feb 18, 2009 10:40 am UTC

Re: Numbers

Postby littlelj » Wed Mar 17, 2010 1:31 pm UTC

xepher wrote:
shoelessschuler wrote:Do I seriously have to drink 9 glasses of water a day? :(

I remember hearing somewhere that about either 33% or 60% comes from food you eat. Well, it depends on the diet. Hence the drastic change in which one it was.

So... not as much as 9, considering the diet.

When you microwave things you find out how much water they have in them.

Bread = lots. Microwaving != toasting.

I liked this comic a lot. I am also in the "where can i buy bras other than scaffolding warehouses?!" category for cup size. :oops:
Dudes, I'm a woman.

yet another steven
Posts: 47
Joined: Wed Jan 13, 2010 10:22 am UTC

Re: Numbers

Postby yet another steven » Wed Mar 17, 2010 1:40 pm UTC

littlelj wrote:I am also in the "where can i buy bras other than scaffolding warehouses?!" category for cup size. :oops:


Well... I like big boobs. You should use a more confident smiley!
Last edited by yet another steven on Wed Mar 17, 2010 2:10 pm UTC, edited 2 times in total.

troggle
Posts: 13
Joined: Wed Jan 20, 2010 5:32 pm UTC

Re: Numbers

Postby troggle » Wed Mar 17, 2010 1:41 pm UTC

littlelj wrote:I am also in the "where can i buy bras other than scaffolding warehouses?!" category for cup size. :oops:


hi a/s/l?

rcox1
Posts: 112
Joined: Wed Nov 11, 2009 11:23 pm UTC

Re: Numbers

Postby rcox1 » Wed Mar 17, 2010 2:28 pm UTC

unfalliblekrutch wrote:the last graph at least makes some sense. The composition of the internet may indeed hit a peak number above 100 because, as a whole, internet users probably are on the higher end of the IQ spectrum compared to non internet users.


The statement kind of disproves itself.

IQ, such as they are, is defined to be a normally distributed by frequency. This means that there is some average which half the people are below and half the people are above. As one moves the same distance in either direction from the average, for instance 10 points, to 90 and 110, the distribution is set up so that the same number of people will have either IQ.

The distribution is also set up so there are precise number of people in a given range about the average. These ranges are called standard deviations, and while the exact numbers vary the upshot is that the the vast majority of the entire population has an IQ below 147.

If the average IQ for the internet were 147, then the curve would be shifted in such a way that the internet would include <1% of the population.(there is some theorem about any measure of a sample of a random variable generating a normal graph, but I am on vacation so am lazy, which is why this is so lame).

If half the population with the lowest IQ did not use the internet, we would be in the 130 range for IQ, no where near to 147. I suspect the 147 number,just like the 9" penis of the E cup, is what people want to here, though I doubt it is a wonderful as some might purport.

yet another steven
Posts: 47
Joined: Wed Jan 13, 2010 10:22 am UTC

Re: Numbers

Postby yet another steven » Wed Mar 17, 2010 2:38 pm UTC

rcox1 wrote:The statement kind of disproves itself.

You're right that it's not right, but it's actually farther from the truth than you give it discredit for:

If half the population with the lowest IQ did not use the internet, we would be in the 130 range for IQ, no where near to 147.


Not in the 130 range: the normal distribution has an exponentially decreasing tail, so if you take a bunch of people, kill all who have an IQ less than 100 and calculate the average IQ for the remaining half, you will find that almost all of them are still pretty close to 100. The number you are after is 100+15 * Phi^{-1)(0.75), where Phi^{-1} is the inverse of the standard normal cdf. The average IQ of the smartest 50% of people is only 110.

Edit: I can't resist elaborating.To get an average IQ of 130, you have to kill everyone with an IQ of 125 or less, leaving only 4.5% of the population alive, half of whom have an IQ between 125 and 130. To get an average IQ of 147, you have to get rid of everyone with an IQ of 144 or less, leaving only the smartest 0.17% of the people, half of whom have an IQ between 144 and 147.

That's a lot of dead people :shock:

cream wobbly
Posts: 170
Joined: Mon Aug 03, 2009 3:07 pm UTC

Re: "Numbers" Discussion

Postby cream wobbly » Wed Mar 17, 2010 3:12 pm UTC

I always knew that the number of glasses of water a day you drink is correlated with penis size in inches.

One glass, one inch. Close enough.

User avatar
squall_line
Posts: 170
Joined: Fri Mar 20, 2009 2:36 am UTC

Re: Numbers

Postby squall_line » Wed Mar 17, 2010 3:25 pm UTC

SheeEttin wrote:The "99 Problems" graph seems to be incomplete. It does show statistics on the amount of problems, but it does not show any data on whether or not a bitch is one.


Best. First. Post. Ever. :D

IMHO, of course.

Oh, and you owe my employer a new keyboard, since keyboards don't like coffee. Luckily, the antique CRT monitor survived the spit-take. :)

User avatar
bmonk
Posts: 662
Joined: Thu Feb 18, 2010 10:14 pm UTC
Location: Schitzoed in the OTT between the 2100s and the late 900s. Hoping for singularity.

Re: "Numbers" Discussion

Postby bmonk » Wed Mar 17, 2010 3:28 pm UTC

I haven't done the internet searches, but according to my email inbox (Viagra, penis enlargement, breast enlargement, drugs, especially for Alzheimer's, and Nigerian heiresses) most of America should have big, stiff penises and huge breasts, but not know what to do with them--oh, and they will soon have 10% of $13.3 million.

Oh, and another thought: for some of these numbers, maybe we should ask, "By what metric?" -- as here: http://spikedmath.com/039.html
Last edited by bmonk on Wed Mar 17, 2010 3:34 pm UTC, edited 1 time in total.
Having become a Wizard on n.p. 2183, the Yellow Piggy retroactively appointed his honorable self a Temporal Wizardly Piggy on n.p.1488, not to be effective until n.p. 2183, thereby avoiding a partial temporal paradox. Since he couldn't afford two philosophical PhDs to rule on the title.

fiz
Posts: 1
Joined: Wed Mar 17, 2010 3:27 pm UTC

Re: "Numbers" Discussion

Postby fiz » Wed Mar 17, 2010 3:28 pm UTC

I have one thing to say to this:

"There are <x> flowers in a vase..."

User avatar
CorruptUser
Posts: 10546
Joined: Fri Nov 06, 2009 10:12 pm UTC

Re: Numbers

Postby CorruptUser » Wed Mar 17, 2010 3:40 pm UTC

Ok, for IQ tests, the standard deviation is 16 for men, 15 for women. Yes this means there are fewer female geniuses, but it also means fewer female idiots.

The tests become unreliable after about 130.

Men are on average 2 points higher than women for spatial IQ, but this can be explained by sampling bias; dumb men die much sooner than dumb women, skewing the graphs. Women score 2 points higher on memory IQ, but this can be attributed to men becoming senile at a younger age.

Yes, races test differently; in the US, the highest being Black African (bet you didn't see THAT coming!), and the lowest being African American, so saying IQ is based entirely on genetics is clearly flawed. That or since most African Americans are in part decended from slave owners, racist southerners have the lowest IQ, but get counted as "white" so they don't show up as the lowest.
Last edited by CorruptUser on Wed Mar 17, 2010 3:52 pm UTC, edited 4 times in total.


Zylon
Posts: 142
Joined: Fri Jan 30, 2009 5:37 pm UTC

Re: "Numbers" Discussion

Postby Zylon » Wed Mar 17, 2010 3:47 pm UTC

Hooray for sampling bias.

Dorque
Posts: 2
Joined: Wed Mar 17, 2010 4:04 pm UTC

Re: "Numbers" Discussion

Postby Dorque » Wed Mar 17, 2010 4:06 pm UTC

147 is the top score you can get on the Facebook IQ Test. Now you know.

yet another steven
Posts: 47
Joined: Wed Jan 13, 2010 10:22 am UTC

Re: Numbers

Postby yet another steven » Wed Mar 17, 2010 4:08 pm UTC

CorruptUser wrote:Yes, races test differently; in the US, the highest being Black African (bet you didn't see THAT coming!)

Cool! I didn't know that.

so saying IQ is based entirely on genetics is clearly flawed.

Another reason this could not be true is because from what I heard, it's actually pretty effective to practice for IQ tests, as many people are doing who get them as part of their job interview. On the other hand, the African Americans might be of quite different genetic descent than Black Africans. Africa is a large; I'm sure slavers did not select Africans uniformly at random from the entire continent. Similarly, I think people who emigrate may be smarter than average. (In my personal experience, I notice that exchange students tend to be brighter than average students.)

I'm sure there are not that many people saying it's entirely based on genetics? That's a strong claim...

acd
Posts: 21
Joined: Fri Jul 11, 2008 2:02 pm UTC

Re: Numbers

Postby acd » Wed Mar 17, 2010 4:19 pm UTC

Panama1984 wrote:So, is it odd that I'm 3 points higher than the internet average in IQ and only an inch off in penis size?



iq 147 and 9" are not average, just the most freqent answers

JustAMinute
Posts: 2
Joined: Wed Mar 17, 2010 4:38 pm UTC

Re: "Numbers" Discussion

Postby JustAMinute » Wed Mar 17, 2010 4:51 pm UTC

Did anyone else get the Star Trek reference in this comic?

yet another steven
Posts: 47
Joined: Wed Jan 13, 2010 10:22 am UTC

Re: "Numbers" Discussion

Postby yet another steven » Wed Mar 17, 2010 4:57 pm UTC

JustAMinute wrote:Did anyone else get the Star Trek reference in this comic?

Fifth post, dude :D

User avatar
JustTheBast
Posts: 10
Joined: Mon Sep 03, 2007 8:00 am UTC
Location: Oberhausen, Germany
Contact:

Re: Numbers

Postby JustTheBast » Wed Mar 17, 2010 5:01 pm UTC

yet another steven wrote:To get an average IQ of 130, you have to kill everyone with an IQ of 125 or less, leaving only 4.5% of the population alive, half of whom have an IQ between 125 and 130.


Sounds like a plan. Let's get started. :twisted:

JustAMinute
Posts: 2
Joined: Wed Mar 17, 2010 4:38 pm UTC

Re: "Numbers" Discussion

Postby JustAMinute » Wed Mar 17, 2010 5:04 pm UTC

yet another steven wrote:Fifth post, dude :D

Pre-freakin-cisely!

Nihiltres
Posts: 23
Joined: Mon Oct 13, 2008 3:39 pm UTC

Re: Numbers

Postby Nihiltres » Wed Mar 17, 2010 5:18 pm UTC

acd wrote:
Panama1984 wrote:So, is it odd that I'm 3 points higher than the internet average in IQ and only an inch off in penis size?



iq 147 and 9" are not average, just the most freqent answers


Adacore wrote:I had an officially administered IQ test when I was about 7 or 8 (I think my school was trying to get extra funding for having a 'gifted' student with the attached special needs budget) that gave me 147 - is this just a really common result for some reason? Perhaps test scoring methods are poorly designed - do they give poor resolution at high and low ranges, or something?


acd is completely correct, and I really want to yell at Randall for getting this wrong. Modal user != average user! For that matter, Modal self-declaration != average self-declaration, and that's ignoring things like satire which would show up as false positives.

Adacore has a point. IQ is measured based on how many people do how well on the test. So suppose you have an IQ test graded on correct/incorrect/unanswered questions. If there's a random chance that you will make a mistake on any given question (say you somehow randomly misread it or something) then you'll start to see error, because how can you determine whether a person made a mistake randomly, or because they're not smart enough? You can minimize the problem by testing more people (getting more sampling data for the distribution of correct/incorrect answers), by making the test longer or testing multiple times (more data so as to minimize the effect of a random error), but the problem remains. In particular, at high IQs, people will tend to get a very, very high percentage of questions right, assuming they're answerable within the general range of human intelligence (e.g. "Which of these odd 100-digit numbers is prime?" is ridiculous), and there are very few of them (pretty much by definition), so it's hard to get enough data to distinguish the smartest people. So as your actual IQ (as measured by a reasonably good test) rises above 140 or so, so does the chance that the actual IQ is meaningless.

arasam22
Posts: 2
Joined: Wed Nov 04, 2009 3:45 am UTC

Re: "Numbers" Discussion

Postby arasam22 » Wed Mar 17, 2010 5:20 pm UTC

THERE... ARE.... FOUR... LIGHTS!!!

awesome.

i love these graph comics. i could look at them forever.

Kalos
Posts: 172
Joined: Thu Jan 31, 2008 6:45 pm UTC

Re: "Numbers" Discussion

Postby Kalos » Wed Mar 17, 2010 5:40 pm UTC

Another series of cheap references in what's obviously another poster-grab. Hopefully with this one he won't pretend to be "Radiohead-style" pricing though.

FlexGunship
Posts: 17
Joined: Wed Mar 03, 2010 1:40 pm UTC

Re: Numbers

Postby FlexGunship » Wed Mar 17, 2010 5:43 pm UTC

unfalliblekrutch wrote:the last graph at least makes some sense. The composition of the internet may indeed hit a peak number above 100 because, as a whole, internet users probably are on the higher end of the IQ spectrum compared to non internet users.


Wow. What internet are you using? Can I join?

User avatar
manmantong2000
Posts: 51
Joined: Sat Sep 26, 2009 8:24 pm UTC
Location: New Jersey, United States of America
Contact:

Re: "Numbers" Discussion

Postby manmantong2000 » Wed Mar 17, 2010 5:57 pm UTC

Is it just me? or other people also found the handwriting in this comic is kinda sloppy? (and also why not color code line?)
"Me no speak English" makes perfect sense.

User avatar
phillipsjk
Posts: 1213
Joined: Wed Nov 05, 2008 4:09 pm UTC
Location: Edmonton AB Canada
Contact:

Re: Numbers

Postby phillipsjk » Wed Mar 17, 2010 6:15 pm UTC

shoelessschuler wrote:Do I seriously have to drink 9 glasses of water a day? :(

No, you are supposed to drink something like 9 cups (250mL) of water a day. Many drinking glasses hold at least 2 cups.
Did you get the number on that truck?

User avatar
Ephemeron
Posts: 282
Joined: Sun Jan 31, 2010 12:39 pm UTC

Re: "Numbers" Discussion

Postby Ephemeron » Wed Mar 17, 2010 6:29 pm UTC

(posted on the next page)
Last edited by Ephemeron on Wed Mar 17, 2010 8:40 pm UTC, edited 18 times in total.


Return to “Individual XKCD Comic Threads”

Who is online

Users browsing this forum: Mikeski and 30 guests