What is your favorite "data analysis" cartoon?

Question

Data analysis cartoons can be useful for many reasons: they help communicate; they show that quantitative people have a sense of humor too; they can instigate good teaching moments; and they can help us remember important principles and lessons.

This is one of my favorites:

XKCD irony about correlation and causation

As a service to those who value this kind of resource, please share your favorite data analysis cartoon. They probably don't need any explanation (if they do, they're probably not good cartoons!) As always, 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.

@sharpie: are jokes out? We obviously don't want the entire site to be humor, but everyone benefits from a little educational humor in small doses. — Shane, Jul 22 '10 at 05:15
@Sharpie, feel free to close or reopen according to your feelings! I agree with Shane, a bit is ok, but not too much. For example, this question already included a funny cartoon. The jokes question not really a funny joke.... — Peter Smit, Jul 22 '10 at 13:58
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. — Shane, Jul 22 '10 at 14:22
Could we clarify the problem with hotlinking referenced in the P.S.? Might need to fix some of these answers, if it's worth worrying about. — Nick Stauner, Feb 20 '14 at 01:14
According to the [tour](https://stats.stackexchange.com/tour), this question should be closed, since it is a question that has "too many possible answers" and since it is "primarily opinion-based". I'm not complaining, just surprised it has stayed open for this long. — Flimm, Dec 09 '14 at 10:29
I decided to recreate this on [serverfault](http://serverfault.com/q/671249/218888). As it seems like a great idea for some levity in an otherwise serious environment. Here's hoping it survives! — Reaces, Feb 25 '15 at 09:32
Data Science analogy to cartoon in OP. Data Scientist: I went to data science bootcamp and learned how to find correlations in big data. Those insights can be converted into big money. Statistician: But many of those correlations are spurious. Correlation does not imply causation. Data Scientist: Don't give me none of that century old statistics mumbo-jumbo. This is big data. That means the data has everything. So by definition, all relationships in the data are correct. I ring the cash register while you snooze and lose, grandpa. — Mark L. Stone, Dec 19 '15 at 22:42
I wanted to vote down because this is not the site for such discussions but this is my favorite data cartoon, I had it in my office for a while. — Amit Keinan, Apr 27 '21 at 19:46

score 242 · Answer 1 · edited Mar 09 '17 at 17:30

242

Was XKCD, so time for Dilbert:

alt text

Source: http://dilbert.com/strip/2001-10-25

edited Mar 09 '17 at 17:30

Community

1

answered Jul 21 '10 at 15:21

40

Did anyone else notice that the tour guide changes colors between the second and third frames? – assumednormal Aug 22 '12 at 20:41
9

[On RANDU](http://en.wikipedia.org/wiki/RANDU): "We guarantee that each number is random individually, but we don't guarantee that more than one of them is random." – Iain Samuel McLean Elder Mar 18 '14 at 15:11
Link not working, was it this one http://dilbert.com/strip/2001-10-25 ? – Tim Jan 27 '15 at 19:27

score 223 · Answer 2 · edited Aug 16 '14 at 17:48

223

Another from XKCD: ... okay, but because you said that, we're breaking up.

Mentioned here and here.

edited Aug 16 '14 at 17:48

Christian

989
1
9
28

answered Aug 30 '10 at 18:02

Shane

11,961
17
71
89

12

You can't read this one without the alt text. it said something like "But because of that we're totally breaking up" – generic_user Mar 10 '14 at 18:16
@generic_user The alt text (mouseover text) reads: "... okay, but because you said that, we're breaking up." – Benjamin Wang Nov 08 '20 at 01:05

csgillespie · Answer 3 · 2015-01-27T20:43:00.943

206

My favourite Dilbert cartoon:

enter image description here

Source: http://dilbert.com/strip/2008-05-07

edited Jan 27 '15 at 20:43

answered Jul 21 '10 at 15:38

csgillespie

11,849
9
56
85

10

Definitively my favorite cartoon about Data Mining – mlwida Dec 01 '10 at 12:24
Link not working. Was it this one: http://dilbert.com/strip/2008-05-07 ? – Tim Jan 27 '15 at 19:27

score 196 · Answer 4 · edited Mar 09 '17 at 17:30

196

One more Dilbert cartoon:

_...

edited Mar 09 '17 at 17:30

Community

1

answered Jul 21 '10 at 15:53

8

This one reminds me of the recent bailout in the States, where they just made up 700 billion number - they said they just wanted a really large number. :) – Roman Luštrik Aug 12 '10 at 08:53
1

Fixed. I had to add some dots after the cartoon since SE didn't allow me to submit the changes :-\ – Ching Chong Feb 05 '15 at 11:30

score 178 · Answer 5 · edited Jun 11 '20 at 14:32

178

One of my favorites from xckd:

Random Number

RFC 1149.5 specifies 4 as the standard IEEE-vetted random number.

edited Jun 11 '20 at 14:32

Community

1

answered Jul 21 '10 at 15:23

Sharpie

4,126
5
21
18

1

But that isn't even prime! – kjetil b halvorsen Jan 25 '17 at 15:27

score 152 · Answer 6 · answered Aug 16 '11 at 15:34

152

Normal Versus Paranormal Distribution

From: A visual comparison of normal and paranormal distributions Matthew Freeman J Epidemiol Community Health 2006;60:6. Lower caption says 'Paranormal Distribution' - no idea why the graphical artifact is occuring.

answered Aug 16 '11 at 15:34

Fomite

21,264
10
78
137

3

I think [this version](http://i.imgur.com/rrjJtoO.png) of the joke works better (from http://www.oneweirdkerneltrick.com), though apparently this version was seven years earlier. – Danica Feb 03 '15 at 01:06
this isn't really funny. it's more of a twist on english terms – Mar 07 '16 at 18:20
1

@zero "A twist on English terms" describes a great many jokes – Fomite Mar 07 '16 at 18:22
Yeah - I think all those jokes suck. There is no underlying statistical humour. This joke should be put on the English stackexchange instead. – Mar 07 '16 at 18:23
@Phil: In the 2-dimensional version linked by Dougal, the "paranormal distribution" is indeed a (bizarrely truncated) distribution; so the joke has some statistical content. It doesn't work in one dimension, where your comment certainly applies. – John Bentin Jun 19 '16 at 17:47

score 150 · Answer 7 · answered Apr 06 '11 at 08:35

150

XKCD: significant

'So, uh, we did the green study again and got no link. It was probably a--' 'RESEARCH CONFLICTED ON GREEN JELLY BEAN/ACNE LINK; MORE STUDY RECOMMENDED!'

xkcd: significant

answered Apr 06 '11 at 08:35

Henrik

13,314
9
63
123

17

This is by far my favorite cartoon of all time. It's super educational. It really gets to the heart of the definition of a p-value. In fact, I bet that less than 10% the students who pass a college freshman "intro to stats" class get this joke, and this makes me sad. – WetlabStudent Jan 15 '14 at 03:36
4

Maybe so! Fortunately for freshmen, @Glen_b has offered [an excellent breakdown here](http://stats.stackexchange.com/a/88067/32036). – Nick Stauner Feb 27 '14 at 01:03
2

Great! But yellow appears twice :P – Rodrigo Mar 01 '16 at 00:51
4

this is a pretty good joke as it clearly demonstrates why repeated multiple testing is dangerous. For anyone interested check out Bonferi correction to deal with this. – Mar 07 '16 at 18:21

score 139 · Answer 8 · edited Dec 15 '12 at 20:29

139

I just came across this and loved it:

alt text

(http://xkcd.com/795/).

edited Dec 15 '12 at 20:29

whuber

281,159
54
637
1,101

answered Sep 24 '10 at 04:09

Tal Galili

19,935
32
133
195

score 128 · Answer 9 · edited Mar 09 '17 at 17:30

128

Image at bp1.blogger.com.

edited Mar 09 '17 at 17:30

Community

1

answered Jul 26 '10 at 19:51

Harvey Motulsky

14,903
11
51
98

8

That's great. The standard way of dealing with outliers. – Shane Jul 26 '10 at 20:06
2

Who's the artist? – blubb Sep 25 '14 at 16:01
1

This cartoon was drawn by [Ben Shabad](http://davidmlane.com/ben/cartoons.html) – ff524 Jan 07 '16 at 20:20

score 122 · Answer 10 · edited Jul 21 '17 at 08:45

122

Another from xkcd #833:

And if you labeled your axes, I could tell you exactly how MUCH better.

edited Jul 21 '17 at 08:45

hongsy

377
3
14

answered Jun 30 '11 at 05:50

raegtin

9,090
12
48
53

score 114 · Answer 11 · edited Nov 29 '10 at 14:53

114

By the third trimester, there will be hundreds of babies inside you.

Also from XKCD

edited Nov 29 '10 at 14:53

Henrik

13,314
9
63
123

answered Nov 27 '10 at 22:27

Adam

582
1
5
7

Neil McGuigan · Answer 12 · 2017-12-21T20:57:28.523

108

This isn't technically a cartoon, but close enough:

edited Dec 21 '17 at 20:57

answered Jul 26 '10 at 23:04

Neil McGuigan

9,292
13
54
62

10

That's definitely my favorite. I always have to stop on this and laugh when scrolling over this page. It's just so bad!! – Henrik Sep 09 '10 at 09:02
2

+1 'Natch, [Randal Munroe has a sweet variation on this](https://xkcd.com/688/). – Alexis Sep 24 '19 at 15:29
Yep: a pie is a pie chart telling you how much pie is left. – Adrian Keister Mar 31 '20 at 21:03

score 101 · Answer 13 · edited May 04 '12 at 22:09

101

Nice. The importance of variance when thinking about a population.

enter image description here

Saturday Morning Breakfast Cereal

edited May 04 '12 at 22:09

whuber

281,159
54
637
1,101

answered Mar 25 '12 at 12:42

Tal Galili

19,935
32
133
195

score 100 · Answer 14 · answered Sep 24 '10 at 04:13

100

this too: alt text

answered Sep 24 '10 at 04:13

Tal Galili

19,935
32
133
195

4

To be honest, those are the bad physicists. The good physicists stick around and make a name for themselves. – Iterator Aug 06 '11 at 03:17
3

Yet it's amazing how often it works... – wnoise Sep 22 '11 at 04:44
Economics defined – luchonacho Oct 18 '21 at 19:40

score 97 · Answer 15 · edited May 04 '12 at 22:21

97

There is this one on Bayesian learning:

alt text

edited May 04 '12 at 22:21

whuber

281,159
54
637
1,101

answered Aug 30 '10 at 23:04

ebony1

2,143
21
13

1

What's the source? – Shane Aug 31 '10 at 15:02
1

It was taken from Mike West's website: http://www.stat.duke.edu/~mw/fineart.html – ebony1 Aug 31 '10 at 16:23

score 93 · Answer 16 · edited Oct 09 '16 at 15:32

93

And another one from xkcd.

Title: Self-Description

alt text

The mouseover text:

The contents of any one panel are dependent on the contents of every panel including itself. The graph of panel dependencies is complete and bidirectional, and each node has a loop. The mouseover text has two hundred and forty-two characters.

edited Oct 09 '16 at 15:32

Community

1

answered Sep 01 '10 at 15:02

Henrik

13,314
9
63
123

score 78 · Answer 17 · edited Aug 28 '12 at 18:38

78

Here is a nice one (the inadequacy about average ratings)

edited Aug 28 '12 at 18:38

gung - Reinstate Monica

132,789
81
357
650

answered Aug 28 '12 at 18:22

Sapsi

101
2
2

Kinda disappointed that Munroe did not work "[selection bias](https://en.wikipedia.org/wiki/Selection_bias)" into the text. – Alexis Sep 24 '19 at 15:32

score 77 · Answer 18 · edited Aug 21 '13 at 01:39

77

Another one from xkcd:

Hell, my eighth grade science class managed to conclusively reject it just based on a classroom experiment. It's pretty sad to hear about million-dollar research teams who can't even manage that.

Alt-text:

Hell, my eighth grade science class managed to conclusively reject it just based on a classroom experiment. It's pretty sad to hear about million-dollar research teams who can't even manage that.

edited Aug 21 '13 at 01:39

naught101

4,973
1
51
85

answered May 06 '11 at 12:13

j.p.

180
4
10

score 72 · Answer 19 · edited Feb 24 '15 at 20:50

72

Here's another one from Dilbert:

enter image description here

edited Feb 24 '15 at 20:50

gung - Reinstate Monica

132,789
81
357
650

answered Jul 21 '10 at 15:43

Shane

11,961
17
71
89

Looks like this one needs an updated image. – Nick Orlando Feb 24 '15 at 20:45

score 68 · Answer 20 · answered Dec 17 '11 at 21:19

68

http://andrewgelman.com/2011/12/suspicious-histograms/

answered Dec 17 '11 at 21:19

zbicyclist

3,363
1
29
34

score 65 · Answer 21 · answered Oct 25 '12 at 13:01

65

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

Tal Galili

19,935
32
133
195

score 63 · Answer 22 · edited Mar 09 '17 at 17:30

63

More about design and power than analysis, but I like this one

alt text

edited Mar 09 '17 at 17:30

Community

1

answered Jul 22 '10 at 13:49

Freya Harrison

3,212
4
25
31

score 56 · Answer 23 · edited Mar 09 '17 at 17:30

56

Source: unknown. Posted on flowingdata.com.

edited Mar 09 '17 at 17:30

Community

1

answered Jun 26 '14 at 11:36

Jot eN

297
7
15

There could be a secondary y-axis labeled "Ballerina chicken with cloaca". – Waldir Leoncio Jan 31 '19 at 09:38

score 55 · Answer 24 · edited May 04 '12 at 22:12

55

enter image description here

Saturday Morning Breakfast Cereal

edited May 04 '12 at 22:12

whuber

281,159
54
637
1,101

answered Aug 05 '11 at 13:32

Richie Cotton

644
9
15

3

yeah but.... this one isn't true... it mostly depends on how you parameterize the time variable $t$... i guess if you go back far enough, but come on... – William Sep 22 '11 at 17:24

score 55 · Answer 25 · answered Dec 19 '11 at 17:41

55

A classic...

Never discuss multiple testing without it

answered Dec 19 '11 at 17:41

guest

2,381
14
11

"Because medical research findings can be difficult to reconcile, are not always pre-digested, and can seem overwhelming to us casual observers, let us make fun of those who dedicate their lives to obtaining them." – rolando2 Feb 09 '14 at 00:53
16

@rolando2 As a medical researcher, I find the sensationalist incompetence of mainstream science reporters hilarious. – Superbest Feb 26 '14 at 12:08
There's a listserv from HealthNewsReview devoted to evaluating media handling of health research findings. – rolando2 Mar 09 '15 at 14:14
4

As a statistician, I find the sensationalist incompetence of mainstream "data analysts" hilarious. – StatsStudent Mar 07 '16 at 21:35

score 54 · Answer 26 · answered May 05 '11 at 09:27

54

enter image description here

Found this one in the comments on Andrew Gelman's blog.

answered May 05 '11 at 09:27

Henrik

13,314
9
63
123

This chart is wrong, correct percentage is about fifty-fifty. – juliohm Mar 10 '14 at 16:09

score 53 · Answer 27 · answered Aug 31 '10 at 15:58

53

I found this from a NoSQL presentation, but the cartoon can be found directly at

http://browsertoolkit.com/fault-tolerance.png

alt text

answered Aug 31 '10 at 15:58

Josh Hemann

3,384
1
21
18

12

Can you please explain this cartoon? – Christian Sep 03 '14 at 00:00

score 47 · Answer 28 · edited Mar 09 '17 at 17:30

47

Allright, I think this one is hilarious- but let's see if it passes the Statistical Analysis Miller test.

Fermirotica

I love how Google handles dimensional analysis. Stats are ballpark and vary wildly by time of day and whether your mom is in town.

edited Mar 09 '17 at 17:30

Community

1

answered Jul 21 '10 at 18:18

Sharpie

4,126
5
21
18

19

Statistical voyeurism? And there we were wondering what to call the site... – walkytalky Jul 23 '10 at 15:48

score 46 · Answer 29 · edited Oct 25 '11 at 20:57

46

From xkcd:

enter image description here This is data analysis in the form of a cartoon, and I find it particularly poignant.

The universe is probably littered with the one-planet graves of cultures which made the sensible economic decision that there's no good reason to go into space--each discovered, studied, and remembered by the ones who made the irrational decision.

edited Oct 25 '11 at 20:57

Brian Diggs

1,021
8
17

answered May 05 '11 at 21:04

Zach

22,308
18
114
158

score 44 · Answer 30 · edited Mar 09 '17 at 17:30

44

Another one from xkcd:

Coconuts are so far down to the left they couldn't be fit on the chart. Ever spent half an hour trying to open a coconut with a rock? Down with coconuts.

edited Mar 09 '17 at 17:30

Community

1

answered Jul 23 '10 at 08:18

Jeromy Anglim

42,044
23
146
250

14

Bananas are always tasty. – Anonymous Type Dec 01 '10 at 04:00
2

@AnonymousType and easy! – tdc Feb 23 '12 at 13:07
Where are persimmons? – EngrStudent May 18 '13 at 03:18
@EngrStudent: they're simultaneously off both ends of the tasty scale. – naught101 Aug 21 '13 at 01:35
Needs replication. – Nick Stauner Jan 17 '14 at 11:51
4

I'm surprised Durian isn't on here... – Frank H. Oct 20 '15 at 17:41
Unsure what to make of the tomatoes ranking low on tasty: perhaps Randall does not know about heirloom varieties? – Alexis Apr 25 '17 at 23:13
I had this one posted on my door for about 7 years :) – Jens Feb 03 '20 at 13:13

Macro · Answer 31 · 2012-05-08T19:33:17.347

39

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

edited May 08 '12 at 19:33

answered May 05 '12 at 01:58

Macro

40,561
8
143
148

2

I like the percentages given in the third panel. – cardinal May 08 '12 at 23:00

score 39 · Answer 32 · edited Nov 29 '10 at 23:10

39

Here's a somewhat more technical one.

alt text

edited Nov 29 '10 at 23:10

chl

50,972
18
205
364

answered Nov 29 '10 at 20:45

score 38 · Answer 33 · edited Mar 09 '17 at 17:30

38

From xkcd:

You don't use science to show that you are right, you use science to become right.

If some people who really believe that everything should be scientifically tested would actually walk their talk than they this comic might even show an event that actually happens.

edited Mar 09 '17 at 17:30

Community

1

answered Jul 22 '10 at 14:32

Christian

989
1
9
28

Jot eN · Answer 34 · 2017-10-25T14:51:58.027

37

Here is a very meaningful chart..

Bizarro

edited Oct 25 '17 at 14:51

answered Mar 18 '14 at 10:15

Jot eN

297
7
15

score 33 · Answer 35 · answered Jan 18 '13 at 19:50

33

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.

answered Jan 18 '13 at 19:50

dimitriy

31,081
5
63
138

Interestingly enough, that suggestion (number of pages, stacks of paper, etc.) is effectively a pseudo log-scale. – Ben Jul 18 '18 at 01:31

score 31 · Answer 36 · edited Jun 11 '20 at 14:32

31

edited Jun 11 '20 at 14:32

Community

1

answered Sep 29 '12 at 23:06

Tomas

5,735
11
52
93

2

This is just one flag for torture never being funny, even indirectly or allusively. – Nick Cox Nov 12 '13 at 18:11
@NickCox, sorry, my english is not enough to understand your sentence.. – Tomas Nov 14 '13 at 17:45
Personal opinion: Torture isn't funny. – Nick Cox Nov 14 '13 at 18:13
6

@NickCox, come on, it's just numbers! :) – Tomas Nov 14 '13 at 18:14
Your cartoon alludes to people suffering. In turn, I say: come on, you should not find that funny. I won't expand on the point. – Nick Cox Nov 14 '13 at 18:17
4

@NickCox, how does my cartoon allude to *people* anyhow? Its about *data*, not people! – Tomas Nov 14 '13 at 21:27
I can only infer that you don't know the meaning of the word "torture". – Nick Cox Nov 14 '13 at 23:16
1

This is also why torture is no good for anyone, including the torturer. – Nick Stauner Jan 17 '14 at 12:03
@NickStauner I don't get this discussion. Is it that Nick Cox was rather sensitive, transfering the joke about suffering data to suffering people? How does your comment relate to this? – Tomas Jan 17 '14 at 12:09
Oh, I wasn't commenting on the conversation, just the cartoon itself... – Nick Stauner Jan 17 '14 at 12:11
4

Ironically, this cartoon uses "torture" in a common, among statisticians, metaphorical sense of "analyze incorrectly". The humor comes from the fact that it just happens to recall a word meaning something else... But the cartoon isn't about torture, it's about abuse of statistical methods. Though, perhaps one might contend that *abuse* is also a serious and weighty matter, and for that matter, so is weight... – Superbest Feb 26 '14 at 12:18

score 30 · Answer 37 · answered May 13 '12 at 02:05

30

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 02:05

Cyan

2,748
17
21

score 30 · Answer 38 · answered Feb 20 '14 at 01:03

30

From tenso GRAPHICS, as claimed on REDDIT.

answered Feb 20 '14 at 01:03

Nick Stauner

11,558
5
47
105

2

I like that it took me a second, and then another second to get the second joke. :) – Alexis Jun 15 '14 at 22:56

Tomas · Answer 39 · 2013-01-17T10:41:19.890

27

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..."

edited Jan 17 '13 at 10:41

answered Sep 22 '11 at 01:48

Tomas

5,735
11
52
93

6

Independence!!! – KH Kim May 12 '12 at 10:28
@KHKim, we all know that, don't ruin a joke :) – Tomas Aug 14 '12 at 19:14
31

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 Aug 28 '12 at 18:52
5

@Zen again, why do you explain this? Even the security check guy in the story understand this, intuitivelly... don't analyze a joke :-) – Tomas Sep 29 '12 at 21:46
6

By the way, as far as I know, the person who first described this anecdote was none other than [Hugo Steinhaus](http://en.wikipedia.org/wiki/Hugo_Steinhaus) (of the Steinhaus-Banach theorem fame) in his "Mathematical Kaleidoscope". – January Nov 09 '12 at 12:50
Thanks @January! Finally some interesting information :-) – Tomas Jan 17 '13 at 10:38
3

Reminds me of Baldrick's Bullet: http://www.youtube.com/watch?v=pKRxX3s3JlM – Bitwise Jul 14 '13 at 01:00
3

It's kind of hard to enjoy a stat joke when it's screaming "bad stat" at the same time! – PatrickT Mar 18 '14 at 04:34
1

This isn't a *cartoon*. It belongs in a different question. – Glen_b May 12 '15 at 04:10

score 27 · Answer 40 · answered Mar 08 '16 at 17:30

27

Not a cartoon but the best way of not being confused about type I and II errors. And very funny IMHO

answered Mar 08 '16 at 17:30

Mamonu

1
2
4

smillig · Answer 41 · 2012-11-09T12:47:06.230

27

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 09 '12 at 12:47

answered Nov 09 '12 at 12:38

smillig

2,336
28
31

2

I am not sure about this one ... the Frequentist Reasoning seems wrong to me but I cannot explain why :(. – mlwida Nov 09 '12 at 13:50
Because of obvious trying to score a cheap point with deliberate misrepresentation, one of the few xkcd comics I don't like. – Momo Nov 09 '12 at 13:52
1

I'm biased towards the Bayesian interpretation, but the frequentist appears to me to be consistent with the standard frequentist interpretation. I hate null hypothesis significance testing, but if you want to go that route, it seems the null hypothesis is that the sun did not explode. If the null hypothesis is true, the chances of observing a "Yes" would indeed be $\frac{1}{36}$ (negligibly higher if you want to be pedantic and include the chance of a machine malfunction). So we've either seen a rare event and the sun is still there or the sun is gone. Many frequentists default to the latter. – Michael McGowan Nov 09 '12 at 14:42
6

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 09 '12 at 14:42
But if we follow the principle of "$H_0$ is what we want do disprove", then $H_0$=Sun did explode => prob is 1-1/36 => cannot reject $H_0$, which does not mean that $H_1$ is true. I guess this was an attempt to cover a physics joke, because the sun cannot become nova. – mlwida Nov 09 '12 at 15:42
8

@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 09 '12 at 15:56
@MichaelMcGowan good point, I agree. – mlwida Nov 09 '12 at 16:14
This is what Andrew Gelman thinks about this comic: http://andrewgelman.com/2012/11/16808/ – boscovich Nov 11 '12 at 09:27
And Larry Wasserman: http://normaldeviate.wordpress.com/2012/11/09/anti-xkcd/ – Momo Nov 11 '12 at 16:56

score 27 · Answer 42 · answered Dec 22 '10 at 23:06

27

From SMBC:

alt text

answered Dec 22 '10 at 23:06

raegtin

9,090
12
48
53

10

Is it churlish to wince when I read "data is"? – Chris Beeley Dec 26 '10 at 23:06
9

Not churlish. Petulant. – rolando2 Apr 06 '11 at 10:28
9

Not Petulant. Literate. ;) – A.M. Aug 15 '13 at 18:50
I wince more that idea that all perspectives are subjectively situated is sad. – Alexis Jun 15 '14 at 22:48
The only time I hear the word "datum" used by *scientists*, it is in the [geodetic](https://en.wikipedia.org/wiki/Geodetic_datum) sense. Perhaps I have a biased sample? – GeoMatt22 Oct 07 '16 at 05:11

score 26 · Answer 43 · edited Mar 09 '17 at 17:30

26

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 09 '17 at 17:30

Community

1

answered Mar 25 '12 at 17:39

conjugateprior

19,431
1
55
83

XKCD link is https://xkcd.com/690/ – Kevin O'Bryant Jan 06 '16 at 20:30
Thanks. Changed the URL to the one recommended for embedding (not the one you suggest). – conjugateprior Jan 06 '16 at 20:41

score 26 · Answer 44 · 2012-05-12T10:12:47.877

26

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

answered May 10 '12 at 20:23

5

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
6

@whuber Thanks for your comment. I added a bit of details in order to clarify its meaning and relationship with statistics. – May 11 '12 at 23:15
2

Is this Avignon bridge? – Tomas Sep 29 '12 at 23:10
@Tomas It looks similar indeed. – Oct 01 '12 at 13:32

score 22 · Answer 45 · answered Feb 20 '15 at 19:47

22

This one may be a little too real for anyone involved in academic research...

If all else fails, use "significant at the p > .05 level" and hope no one notices."

See the original here.

answered Feb 20 '15 at 19:47

Richard Border

1,128
9
26

1

I like this XKCD, but Randall Munroe's implication about the value of p = exactly .050 is incorrect. http://stats.stackexchange.com/questions/60825/is-a-p-value-of-0-04993-enough-to-reject-null-hypothesis explains why this is the case. – user1205901 - Reinstate Monica Apr 14 '15 at 13:49
The difference between incorrect and correct here has measure zero :) – Richard Border May 07 '21 at 17:56

score 22 · Answer 46 · answered Mar 10 '14 at 15:36

22

Because It's PIE, make me laught LOL. hahaha http://portal-statistik.blogspot.com enter image description here

answered Mar 10 '14 at 15:36

user41659

1
1
2

score 21 · Answer 47 · answered Aug 07 '15 at 22:00

21

This one is a hit :)! I've seen it a few days ago.

answered Aug 07 '15 at 22:00

Marcin Kosiński

819
3
12
25

score 17 · Answer 48 · answered Oct 20 '13 at 23:10

17

pie chart example

an 'easy to digest' pie chart example for Rick Astley fans that my students seem to enjoy

answered Oct 20 '13 at 23:10

Theresa

1
1
2

13

Are those mutually exclusive events? :-) – cardinal Oct 21 '13 at 01:32
Reminds me of the "Hey Jude" flowchart: https://flowingdata.com/2011/01/21/hey-jude-flowchart/ – Gaurav Jan 25 '17 at 22:00
Using very similar shades of colour for different responses - naughty boy! – Ben Jul 18 '18 at 01:36

score 16 · Answer 49 · answered Aug 24 '15 at 20:58

16

Sorry it is in Dutch! Translation

Greetings new recruits! Welcome to the training camp for gladiators
Be warned! One in three of you does not survive the training
33.3% ... That's not so bad

answered Aug 24 '15 at 20:58

StijnDeVuyst

2,161
11
17

As a pointer to why we should think about *conditional* probabilities (and from then on, to why tools like logistic regression are useful for predicting risk) this is an excellent cartoon. – Silverfish Dec 19 '15 at 22:39

score 16 · Answer 50 · answered Jan 24 '12 at 14:03

16

answered Jan 24 '12 at 14:03

score 16 · Answer 51 · edited Oct 25 '11 at 20:59

16

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

Brian Diggs

1,021
8
17

answered Nov 27 '10 at 18:14

Tal Galili

19,935
32
133
195

22

?? 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 04 '12 at 22:11

score 15 · Answer 52 · answered Mar 16 '16 at 19:56

15

Quite a trend (I am the one on the left with the laptop)

answered Mar 16 '16 at 19:56

Laurent Duval

2,077
1
20
33

score 15 · Answer 53 · edited Mar 09 '17 at 17:30

15

I wonder if it's OK to use %-points as an abbreviation of percentage points.

http://xkcd.com/985/

percentage points

edited Mar 09 '17 at 17:30

Community

1

answered Jan 29 '13 at 14:18

Marloes

740
6
21

score 14 · Answer 54 · edited Mar 09 '17 at 17:31

14

From XKCD:

Though 100 years is longer than a lot of our resources.

edited Mar 09 '17 at 17:31

Community

1

answered Nov 29 '15 at 16:20

LegionMammal978

101
1
3

score 14 · Answer 55 · edited May 04 '12 at 22:06

14

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 04 '12 at 22:06

whuber

281,159
54
637
1,101

answered May 04 '12 at 21:14

Michael R. Chernick

39,640
28
74
143

1

Is it this one, Michael? – whuber May 04 '12 at 22:06
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 R. Chernick May 07 '12 at 21:06
Being familiar with this book, it was easy to look it up on the Web. I quickly found one of those versions (on a reseller page) that give you access to selected pages. This image is from such a page, so no scanning (on my part) was necessary. – whuber May 07 '12 at 21:11
Okay, sometimes that is possible but I think more often scanning would be necessary. – Michael R. Chernick May 07 '12 at 21:50

score 13 · Answer 56 · answered Aug 26 '15 at 07:03

13

I liked this one. Found on this page.

answered Aug 26 '15 at 07:03

Darshan Chaudhary

103
1
8

score 13 · Answer 57 · edited Mar 16 '16 at 21:17

13

This one makes you think about the importance of thinking about conditional probabilities. Now I don't know what to make of the twist at the end.

edited Mar 16 '16 at 21:17

whuber

281,159
54
637
1,101

answered Oct 28 '15 at 21:14

Waldir Leoncio

2,137
6
28
42

score 12 · Answer 58 · edited May 04 '12 at 22:13

12

enter image description here

John Deering, Strange Brew

edited May 04 '12 at 22:13

whuber

281,159
54
637
1,101

answered Jun 30 '11 at 02:48

zbicyclist

3,363
1
29
34

13

I admit, I don't get it. – mlwida Jun 30 '11 at 07:27
7

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
8

Let epsilon be less than 0 – Dason Jan 14 '14 at 03:00

score 11 · Answer 59 · answered Feb 03 '15 at 00:54

11

Not exactly data analysis but I had a chuckle.

enter image description here

answered Feb 03 '15 at 00:54

Luca

4,410
3
30
52

score 11 · Answer 60 · answered Dec 19 '15 at 22:21

11

Yes, let's not jump to random forest before we work on some simpler branches!

http://www.gocomics.com/andertoons/2015/12/17

answered Dec 19 '15 at 22:21

zbicyclist

3,363
1
29
34

score 10 · Answer 61 · answered Aug 05 '15 at 14:02

10

My favorite is Sidney Harris he has many great cartoons enter image description here

answered Aug 05 '15 at 14:02

Acoustesh

136
1
5

score 10 · Answer 62 · answered Dec 09 '15 at 00:47

10

This one might be useful when introducing the concept of experimental and control groups.

http://www.gocomics.com/looseparts/2015/12/08

answered Dec 09 '15 at 00:47

zbicyclist

3,363
1
29
34

score 10 · Answer 63 · answered Aug 19 '16 at 16:54

10

Life...

"Hearing something a hundred times isn't better than seeing it once"

answered Aug 19 '16 at 16:54

Carlos Medina

1
1
2

score 10 · Answer 64 · answered Mar 28 '12 at 01:07

10

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

answered Mar 28 '12 at 01:07

zbicyclist

3,363
1
29
34

score 9 · Answer 65 · answered Sep 20 '18 at 18:27

9

"Curve-Fitting Methods and the Messages They Send" by xkcd: https://xkcd.com/2048/

answered Sep 20 '18 at 18:27

T.E.G.

1,676
6
16
29

score 8 · Answer 66 · answered Oct 12 '11 at 01:14

8

http://www.gocomics.com/strangebrew/2011/10/11/

enter image description here

answered Oct 12 '11 at 01:14

zbicyclist

3,363
1
29
34

score 8 · Answer 67 · answered Aug 31 '10 at 17:41

8

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.

Gorbachev says to Bush: "that's a very nice golfcart, Mr. President. Can it change how statistics is practiced?" etc. hahahah.

answered Aug 31 '10 at 17:41

shabbychef

10,388
7
50
93

score 7 · Answer 68 · edited Mar 16 '16 at 21:19

7

Note: this is from SMBC (Saturday Morning Breakfast Cereal) by Zach Weiner.

enter image description here

edited Mar 16 '16 at 21:19

whuber

281,159
54
637
1,101

answered May 12 '15 at 03:21

zbicyclist

3,363
1
29
34

score 6 · Answer 69 · answered Jun 27 '16 at 08:15

6

Correlation does not imply causation!

answered Jun 27 '16 at 08:15

Kostia

1,567
1
8
9

score 6 · Answer 70 · answered Mar 14 '18 at 22:45

6

A new one from XKCD, suggesting a preference for a particular plot type:

https://xkcd.com/1967/

hover text: Strictly speaking, 'violin' refers to the internal structure of the data. The external portion visible in the plot is called the 'viola.'

answered Mar 14 '18 at 22:45

Bryan Krause

1,414
6
15

score 5 · Answer 71 · answered Feb 06 '18 at 13:26

5

Statisticians aren't easily cowed.

answered Feb 06 '18 at 13:26

zbicyclist

3,363
1
29
34

score 4 · Answer 72 · answered Jun 15 '14 at 21:05

4

Source: http://www.gocomics.com/andertoons/2014/06/15#.U54J7iigS8A by Mark Anderson, June 15, 2014.

enter image description here

answered Jun 15 '14 at 21:05

zbicyclist

3,363
1
29
34

score 4 · Answer 73 · answered Aug 09 '11 at 14:53

4

answered Aug 09 '11 at 14:53

zbicyclist

3,363
1
29
34

24

Shouldn't it be x-variable? – gung - Reinstate Monica Mar 25 '12 at 14:34

score 4 · Answer 74 · answered Nov 12 '13 at 14:38

4

More of a math cartoon than a data analysis cartoon, but also one that makes you think a bit.

http://www.gocomics.com/barneyandclyde/2013/11/12/?view=full#.UoI73-KQOfs enter image description here

answered Nov 12 '13 at 14:38

zbicyclist

3,363
1
29
34

DaL · Answer 75 · 2015-11-18T13:12:29.283

3

Overfitting -explanation in a picture (original cartoon)

edited Nov 18 '15 at 13:12

answered Nov 18 '15 at 12:03

DaL

4,462
3
16
27

2

Giving a source would be good practice. Pity about the typo (split infinitives are acceptable to me). – Nick Cox Nov 18 '15 at 12:31
I created the cartoon. Got that result, found it amazing and added the text. What is the typo? How would you phrase the titles? – DaL Nov 18 '15 at 12:39
Fine; so you are fully entitled to claim "(original cartoon)". You fixed the typo I saw (allways for always). – Nick Cox Nov 18 '15 at 12:54

score 3 · Answer 76 · answered Nov 23 '19 at 17:03

3

Gahan Wilson died this week; having presented many a confusing graph myself, I can relate to this:

answered Nov 23 '19 at 17:03

zbicyclist

3,363
1
29
34

score 2 · Answer 77 · answered Dec 28 '16 at 05:57

2

Draw your own data , found it in the below link

[New Year] : http://robertgrantstats.co.uk/drawmydata.html

answered Dec 28 '16 at 05:57

GD_N

193
8

score 2 · Answer 78 · answered Dec 29 '16 at 00:55

True if $P=NP$

True if $P \ne NP$

This is great one about solving NP-complete problems. They come up a lot on the job, like efficient scheduling or how to select the optimal configuration among a number of various options for which you have to search through them all to find the best one.

Think about it anytime you need to cop out of something difficult at work!

score 1 · Answer 79 · answered Jan 10 '18 at 19:49

1

Loose Parts by Dave Blazek 1/10/2018

answered Jan 10 '18 at 19:49

zbicyclist

3,363
1
29
34

score 0 · Answer 80 · answered Mar 11 '21 at 03:47

0

The cartoon can actually be found here https://thejenkinscomic.wordpress.com/

answered Mar 11 '21 at 03:47

POC

346
1
8
23

What is your favorite "data analysis" cartoon?

80 Answers80

Random Number

Fermirotica

Linked