What is your favorite “data analysis” cartoon?

Question

This is one of my favorites:

alt text

One entry per answer. This is in the vein of the Stack Overflow question What’s your favorite “programmer” cartoon?.

P.S. Do not hotlink the cartoon without the site's permission please.

I do have to ask though- how come cartoons are in and jokes are out?
These cartoons are useful too; they can be included in a lecture on a particular topic where you are trying to explain a concept (e.g. correlation/causation above). A little humor can help to keep an audience engaged.

3 revs · Answer 1 · 2012-05-08 19:33:17Z

up vote 11 down vote

I don't think this one was posted yet... enter image description here

edited May 8 '12 at 19:33

community wiki

3 revs
Macro

show 1 more comment

Tal Galili · Answer 2 · 2012-10-25 13:01:56Z

up vote 11 down vote

I liked this one:

enter image description here

This is probably fun to show in class as well...

answered Oct 25 '12 at 13:01

community wiki

Tal Galili

raegtin · Answer 3 · 2010-12-22 23:06:03Z

up vote 10 down vote

From SMBC:

alt text

answered Dec 22 '10 at 23:06

community wiki

raegtin

3

Is it churlish to wince when I read "data is"? – Chris Beeley Dec 26 '10 at 23:06

4

Not churlish. Petulant. – rolando2 Apr 6 '11 at 10:28

2 revs, 2 users 71% · Answer 4 · 2011-10-25 20:59:20Z

up vote 9 down vote

From xkcd:

Almost a Chi square...

alt text

As the CoKF approaches 0, productivity goes negative as you pull OTHER people into chair-spinning contests.

edited Oct 25 '11 at 20:59

community wiki

2 revs, 2 users 71%
Tal Galili

4

?? This seems to have nothing to do with stats. (The curve is modeled after energy potentials in physics, not after anything in stats.) – whuber♦ May 4 '12 at 22:11

5 revs user10525 · Answer 5 · 2012-05-12 10:12:47Z

up vote 9 down vote

enter image description here

"The bridge of life"

I took this image from here. This is a "Painting commissioned by Karl Pearson", see. It is considered as a predecessor of the hazard function.

The 'Death' attempts to kill you at different ages using different sorts of weapons which are related to the "failure probability" at the corresponding age.

edited May 12 '12 at 10:12

community wiki

5 revs
user10525

1

It would help dense people like me to see some brief explanation of how this is specifically related to data analysis. Also, please acknowledge (or at least link to) the source: give credit where credit is due. – whuber♦ May 11 '12 at 19:48

3

@whuber Thanks for your comment. I added a bit of details in order to clarify its meaning and relationship with statistics. – user10525 May 11 '12 at 23:15

1

Is this Avignon bridge? – Tomas Sep 29 '12 at 23:10

show 1 more comment

2 revs · Answer 6 · 2013-01-17 10:41:19Z

This is not a cartoon, but a joke worth mentioning:

A statistic professor travels to a conference by plane. When he passes the security check, they discover a bomb in his carry-on-baggage. Of course, he is hauled off immediately for interrogation.

"I don't understand it!" the interrogating officer exclaims. "You're an accomplished professional, a caring family man, a pillar of your parish - and now you want to destroy that all by blowing up an airplane!"

"Sorry", the professor interrupts him. "I had never intended to blow up the plane."

"So, for what reason else did you try to bring a bomb on board?!"

"Let me explain. Statistics shows that the probability of a bomb being on an airplane is 1/1000. That's quite high if you think about it - so high that I wouldn't have any peace of mind on a flight."

"And what does this have to do with you bringing a bomb on board of a plane?"

"You see, since the probability of one bomb being on my plane is 1/1000, the chance that there are two bombs is 1/1000000. This way I am much safer..."

But if you know that "there is a bomb" (yours) in the plane, which we may call event $A$, and you are willing to accept that the "existence of a second bomb" (event $B$) is independent of $A$, then $P(B\mid A)=P(B)=1/1000$. Always condition on what you know. And yeah, I deserve a $-1$ for screwing a good joke.
@Zen again, why do you explain this? Even the security check guy in the story understand this, intuitivelly... don't analyze a joke :-)
By the way, as far as I know, the person who first described this anecdote was none other than Hugo Steinhaus (of the Steinhaus-Banach theorem fame) in his "Mathematical Kaleidoscope".

mbq · Answer 7 · 2012-01-24 14:03:02Z

up vote 6 down vote

answered Jan 24 '12 at 14:03

community wiki

mbq

2 revs · Answer 8 · 2012-03-25 17:45:02Z

up vote 6 down vote

Explaining Away

Since these are a rather sampling theoretic set of cartoons so far, here's one for the Bayesians. (Actually I set it as a class question last year.)

Explaining Away

edited Mar 25 '12 at 17:45

community wiki

2 revs
conjugateprior

2 revs, 2 users 75% · Answer 9 · 2012-05-04 22:06:12Z

up vote 6 down vote

No one put up a cartoon from the cartoon guide to statistics. I like many of them from there and I used a number of them in one of my books. The one that seems to get the most laughs when I use it in a lecture is the one with the statistician going out on a first date. Their comments and thoughts about the making decisions on the menu with the statistician assessing probabilities and the woman just choosing what she likes makes it really hilarious.

enter image description here

edited May 4 '12 at 22:06

community wiki

2 revs, 2 users 75%
whuber

1

Yes. Thanks a lot Bill. I didn't have any idea how I could paste it in. Is uppose I could have scanned it in to a file and then tried pasting it. That would have been a lot of trouble. Is that what you did? There are a few more scenes in that one that are also pretty funny. But this gets the idea across. – Michael Chernick May 7 '12 at 21:06

show 3 more comments

Cyan · Answer 10 · 2012-05-13 02:05:17Z

up vote 6 down vote

enter image description here

Saturday Morning Breakfast Cereal

And the votey (a sort of black-and-white epilogue unique to SMBC):

enter image description here

answered May 13 '12 at 2:05

community wiki

Cyan

Tomas · Answer 11 · 2012-09-29 23:06:37Z

up vote 6 down vote

enter image description here

body must be at least 30 characters; you entered 4 body must be at least 30 characters; you entered 4

answered Sep 29 '12 at 23:06

community wiki

Tomas

2 revs · Answer 12 · 2012-11-09 12:47:06Z

up vote 6 down vote

A Frequentists vs. Bayesians cartoon from XKCD!

Mouse-hover transcript:

'Detector! What would the Bayesian statistician say if I asked whether the--' [roll] 'I AM A NEUTRINO DETECTOR, NOT A LABYRINTH GUARD. SERIOUSLY, DID YOUR BRAIN FALL OUT?' [roll] '... Yes.'

edited Nov 9 '12 at 12:47

community wiki

2 revs
smillig

1

I am not sure about this one ... the Frequentist Reasoning seems wrong to me but I cannot explain why :(. – steffen Nov 9 '12 at 13:50

2

Of course, using a threshold of $p < 0.05$ is ridiculous in this case, but unfortunately many frequentists don't think about other thresholds. – Michael McGowan Nov 9 '12 at 14:42

1

@steffen I would contend that using "$H_0$ = Sun did explode" is not really appropriate in this case. The null hypothesis is supposed to be the default position, so unless I have an incredibly strong reason to believe otherwise, my default will be that the sun did not explode. – Michael McGowan Nov 9 '12 at 15:56

show 6 more comments

2 revs, 2 users 86% · Answer 13 · 2012-05-04 22:13:33Z

up vote 5 down vote

enter image description here

John Deering, Strange Brew

edited May 4 '12 at 22:13

community wiki

2 revs, 2 users 86%
zbicyclist

11

I admit, I don't get it. – steffen Jun 30 '11 at 7:27

1

I think it's that for any other type of presentation, you'd have started by telling a joke. But since mathematicians (or statisticians, here) only think and speak in terms of formulas, this was their (still lame) joke-analogue for opening a presentation. – AdamO Dec 17 '11 at 22:15

Dimitriy V. Masterov · Answer 14 · 2013-01-18 19:50:12Z

Another one from xkcd:

enter image description here

Hover Text:

Knuth Paper-Stack Notation: Write down the number on pages. Stack them. If the stack is too tall to fit in the room, write down the number of pages it would take to write down the number. THAT number won't fit in the room? Repeat. When a stack fits, write the number of iterations on a card. Pin it to the stack.

zbicyclist · Answer 15 · 2012-03-28 01:07:46Z

up vote 3 down vote

http://www.gocomics.com/frazz/2012/03/27/ enter image description here

answered Mar 28 '12 at 1:07

community wiki

zbicyclist

shabbychef · Answer 16 · 2010-08-31 17:41:55Z

up vote 2 down vote

Bush and Gorbachev in a statistical golf cart My favorite was created by Emanuel Parzen, appearing in IMA preprint 663, but this illustrates my degenerate sense of humor.