2303: Error Types

Explain xkcd: It's 'cause you're dumb.
Revision as of 23:30, 6 May 2020 by 162.158.62.221 (talk) (Explanations of type V and VI. Idk how to link a specific section of a Wikipedia article)
Jump to: navigation, search
Error Types
Type IIII error: Mistaking tally marks for Roman neumerals
Title text: Type IIII error: Mistaking tally marks for Roman neumerals

Explanation

Ambox notice.png This explanation may be incomplete or incorrect: Created by a TYPE IX DROID. Please mention here why this explanation isn't complete. Do NOT delete this tag too soon.
If you can address this issue, please edit the page! Thanks.

The comic is riffing on Type I and type II errors. The first two rows of the comic's table are correct definitions for established terms in statistics. Further rows contain suggestions for new terminology.

Explanation of error types
Type Description Explanation
Type I False positive A false positive is a result that indicates a correlation, when there is no correlation in reality.
Type II False negative A false negative is a result that indicates no correlation, when there is a correlation in reality.
Type III True positive for incorrect reasons Type III error is a nonstandard term meant to build off the notion of type I and II errors. Randall's explanations of this and of Type IV errors line up with some definitions of them, but others have been proposed. None have yet been widely adopted.
Type IV True negative for incorrect reasons As with "Type III", this definition is nonstandard and often a bit tongue-in-cheek.
Type V Incorrect result which leads you to a correct conclusion due to unrelated errors Here we get into errors entirely made up by Randall. The idea behind this one is that a botched statistical test might accidentally result in a true conclusion due to completely unrelated errors in the other direction--perhaps during data collection or aggregation.
Type VI Correct result which you interpret wrong This has actually been proposed as a definition of a Type IV error by Marascuilo and Levin.
Type VII Incorrect result which produces a cool graph
Type VIII Incorrect result which sparks further research and the development of new tools which reveal the flaw in the original results while producing novel correct results
Type IX The Rise of Skywalker Star Wars: The Rise of Skywalker is the ninth and final film in the Star Wars Skywalker saga. It received far less critical acclaim than the previous two films in the sequel trilogy.
Type IIII Mistaking tally marks for Roman neumerals [sic] Title text.

Transcript

Ambox notice.png This transcript is incomplete. Please help editing it! Thanks.
  • Type I Error: False positive
  • Type II Error: False negative
  • Type III Error: True positive for incorrect reasons
  • Type IV Error: True negative for incorrect reasons
  • Type V Error: Incorrect result which leads you to a correct conclusion due to unrelated errors
  • Type VI Error: Correct result which you interpret wrong
  • Type VII Error: Incorrect result which produces a cool graph
  • Type VIII Error: Incorrect result which sparks further research and the development of new tools which reveal the flaw in the original results while producing novel correct results
  • Type IX Error: The Rise of Skywalker


comment.png add a comment! ⋅ comment.png add a topic (use sparingly)! ⋅ Icons-mini-action refresh blue.gif refresh comments!

Discussion

As of the time of this post, the title text is "Type IIII error: Mistaking tally marks for Roman neumerals". Is "neumerals" a typo, or is there a joke in there that I'm missing? Cosmogoblin (talk) 22:52, 6 May 2020 (UTC)

Looks like a typo to me. Randall's patrons should have caught this for him!
ProphetZarquon (talk) 23:07, 6 May 2020 (UTC)
That particular mistake is actually just called a type error.
162.158.62.221 23:12, 6 May 2020 (UTC)
"Not sure if humorous or just wrong..." to paraphrase a Fry meme... It does appear to be a typographical error; Though I suppose it could be a misspelling?
ProphetZarquon (talk) 17:46, 9 May 2020 (UTC)
Or, surely, a Type-0. 162.158.159.82 01:09, 7 May 2020 (UTC)
I am Type-O and as a Type-O Negative, I'm VERY popular at the blood bank! (Universal Donor) 108.162.216.48 01:25, 7 May 2020 (UTC)
Actually, that's an over-simplification. Type-O-neg is only universal for whole blood donations, and only truly universal for whole blood if you are also CMV-negative. For plasma donations, Type-AB (both -pos and -neg) are universal donors. For platelets, only Type-AB-pos is universal donor. 162.158.74.81 16:11, 7 May 2020 (UTC)
When I play Type-O Negative at the blood bank I get asked to turn it down. I'm sure it's loud enough to wake the dead, but the vampires should be glad I woke them in time to make a withdrawal.</badjokes>
Side note, my last name starts with O, so I sign high-score boards with 00O.
ProphetZarquon (talk) 17:46, 9 May 2020 (UTC)
I was hoping it could be bent to be a tribute to John von Neumann. 172.68.189.223 05:00, 7 May 2020 (UTC)
Looks like same type of typo he made at word "blag": [Intentional.] 172.69.54.189 08:04, 7 May 2020 (UTC)
Perhaps a reference to Neumes? 162.158.158.211 09:47, 7 May 2020 (UTC)
But according to [1530] this is an impossible typo to make ?!? Gunterkoenigsmann (talk) 07:27, 24 May 2020 (UTC)

108.162.216.48 01:24, 7 May 2020 (UTC) The explanation builds on definitions of terms in statistics. That's fine, but there are also non-statistical usages, just for example whether someone has now (or had before) the COVID-19 virus. A false positive is a test result which incorrectly indicates that a particular condition or attribute is present, and a false negative is a test result which incorrectly indicates that a particular condition or attribute is absent. A particular test is useful when its incidence of Type I and II errors is low. Types III and IV in that context would be given by poorly designed tests which, even if they give correct results, do it for unsupportable reasons and are therefore unreliable for future results. Types V, VI, VII, and VIII are necessary fillers in the sequence, once you decide that calling The Rise of Skywalker a mistake has to be error type IX simply because it's the ninth film in the series. JohnB (talk) 00:54, 7 May 2020 (UTC)

I could convince myself that the errors are a reference to each Star Wars movie. Definitely a stretch but I'd believe it. 162.158.106.24 02:00, 7 May 2020 (UTC)

Surely an Error of the Third Kind is when an Alien gets lost and lands on Earth. Arachrah (talk) 11:28, 7 May 2020 (UTC)

Can we please stop categorizing every comic about scientific research and methods as COVID-19 related? This is getting quite silly.162.158.187.201 14:31, 7 May 2020 (UTC)

While some members of the public this year are hearing about false positive and false negative for the first time in connection with SARS-COV2 and swab testing, usually in some counterfactual argument that getting tested now increases the probability that you got infected last week or something - I don't see that coronavirus needs to be mentioned in an explanation of the comic, given that it's xkcd and science happens here all the time. If this script ran in Arlo & Janis then a justification like that would be required... although today (May 7th) they're discussing Pavlov's dogs... is there a web page that explains Arlo & Janis? Robert Carnegie [email protected] 162.158.158.225 01:26, 8 May 2020 (UTC)

I spent some time musing about whether the Skywalker saga could be taken as an exemplar of each of the types (e.g. the hypothesis for #1 is that Anakin Skywalker could bring balance to the Force, and the experiment was assigning him a tutor), but it's a stretch. I do like the idea that #9 is the epitome of errordom.16:07, 7 May 2020 (UTC)

I suspect that people who like Episode IX would disagree, but I haven't found evidence that they even exist, so I can't say for sure. GreatWyrmGold (talk) 22:02, 7 May 2020 (UTC)

I'm not sure where this would fit in, but a correct answer that is viewed by the experimenter as being incorrect because of a misunderstanding, so that he changes the recording of the measurement, resulting in an incorrect measurement that he view as correct. This happened in a college classroom exercise involving Reynolds numbers. Above a certain value of Reynold's number, laminar flow will change to turbulent flow. However, that number is not where the change occurs but where laminar flow becomes unstable. One student changed all the measurements to indicate that the flow changed to turbulent almost immediately. I wondered what could have sped up the transition, thinking of things like loud noises and vibrations affecting the apparatus. The student immediately and loudly yelled that he "hadn't changed the numbers", with the vehemence indicating that he had actually changed the numbers. One problem with analog meters was that some people staring at the meters actually thought that the needle moved when it actually hadn't. Digital meters with automatic logging tended to get rid of this problem. 17:17, 7 May 2020 (UTC)

Everyone knows that of all the Reynolds numbers, only two actually matter. Burt and Debbie. (You're gonna say Ryan as well, I suppose...) 162.158.159.82 00:12, 8 May 2020 (UTC)

Wikipedia on Roman Numerals says: at some early time the Romans started to use the shorter forms IV and IX. But originally their numbers included 4 as IIII and 9 as VIIII. I wasn't sure if the Roman Empire used IV and IX at all, and I still am not sure, but if the change came early, so called, then I guess so. I wonder if there's a particular issue of stone cracking if you try to carve IIII on it, or it takes longer. Robert Carnegie [email protected] 162.158.159.74 01:36, 8 May 2020 (UTC)

IIII is a clock error.172.69.63.173 23:13, 8 May 2020 (UTC)

I notice in the wikipedia article for type III errors, which mentions several people's proposed definitions, some of them mention one not listed here: a correct answer, but they were investigating the wrong question to begin with.--162.158.74.81 17:58, 9 May 2020 (UTC)

Roman numerals are tally marks. Duh. —Kazvorpal (talk) 20:09, 16 May 2020 (UTC)