Difference between revisions of "1999: Selection Effect"

Explain xkcd: It's 'cause you're dumb.
Jump to: navigation, search
(Explanation)
(Explanation: informed consent, ethics, kidnapping)
Line 14: Line 14:
 
The title refers to the effect in scientific fields where the people who agree to work with a researcher can affect the outcome.  For example if I said I wanted to do a study on an embarrassing condition, people who know they have it might be more apprehensive and thus not participate.  This can skew the results to say that the condition is rarer than it is. This is called the {{w|selection bias}}, or more precisely, the {{w|self-selection bias}}. Ironically, to avoid any selection bias, researchers would have to force their ''randomly'' selected subjects to participate in their study, but yet the uneasiness shown by the mandatory subjects again skews the results.
 
The title refers to the effect in scientific fields where the people who agree to work with a researcher can affect the outcome.  For example if I said I wanted to do a study on an embarrassing condition, people who know they have it might be more apprehensive and thus not participate.  This can skew the results to say that the condition is rarer than it is. This is called the {{w|selection bias}}, or more precisely, the {{w|self-selection bias}}. Ironically, to avoid any selection bias, researchers would have to force their ''randomly'' selected subjects to participate in their study, but yet the uneasiness shown by the mandatory subjects again skews the results.
  
Ponytail says that people who agree to be in a study at their lab are less likely to attempt to escape. This makes sense, since if you agreed to the study, you know why you are there, while if you didn't, you may have been kidnapped. This implies that Ponytail has recently kidnapped people for a study, and that most of the people she kidnapped called the police, as one should do when being kidnapped{{Citation needed}}. It is not clear how many people who did agree to participate may have attempted to call the police for assistance regardless; compare the [https://en.wikipedia.org/wiki/Stanford_prison_experiment Stanford Prison Experiment].
+
Ponytail says that people who agree to be in a study at their lab are less likely to attempt to escape. This makes sense, since if you agreed to the study, you know why you are there, while if you didn't, you may have been kidnapped. This implies that Ponytail has recently kidnapped people for a study, and that most of the people she kidnapped called the police, as one should do when being kidnapped{{Citation needed}}. The comic incorrectly shows Ponytail being allowed to present the results of this study at a conference; reputable scientific journals and conferences should not legitimize studies that clearly violate their ethical norms, such as by failing to obtain informed consent from human subjects before experimenting on them.  It is not clear how many people who did agree to participate may have attempted to call the police for assistance regardless; compare the [https://en.wikipedia.org/wiki/Stanford_prison_experiment Stanford Prison Experiment].
  
 
The title text says that people who didn't agree to go inside an MRI machine are more likely to escape the machine mid-scan. Again, this makes sense, because being put somewhere against your will makes you unsure what is going on, making you try to find a way out.
 
The title text says that people who didn't agree to go inside an MRI machine are more likely to escape the machine mid-scan. Again, this makes sense, because being put somewhere against your will makes you unsure what is going on, making you try to find a way out.

Revision as of 03:53, 29 May 2018

"1999", this comic's number, redirects here. For the comic named "1999", see 855: 1999.
Selection Effect
fMRI testing showed that subjects who don't agree to participate are much more likely to escape from the machine mid-scan.
Title text: fMRI testing showed that subjects who don't agree to participate are much more likely to escape from the machine mid-scan.

Explanation

Ambox notice.png This explanation may be incomplete or incorrect: edited by an fMRI machine escapee! (help! get me out of here) - Please change this comment when editing this page. Do NOT delete this tag too soon.
If you can address this issue, please edit the page! Thanks.

The title refers to the effect in scientific fields where the people who agree to work with a researcher can affect the outcome. For example if I said I wanted to do a study on an embarrassing condition, people who know they have it might be more apprehensive and thus not participate. This can skew the results to say that the condition is rarer than it is. This is called the selection bias, or more precisely, the self-selection bias. Ironically, to avoid any selection bias, researchers would have to force their randomly selected subjects to participate in their study, but yet the uneasiness shown by the mandatory subjects again skews the results.

Ponytail says that people who agree to be in a study at their lab are less likely to attempt to escape. This makes sense, since if you agreed to the study, you know why you are there, while if you didn't, you may have been kidnapped. This implies that Ponytail has recently kidnapped people for a study, and that most of the people she kidnapped called the police, as one should do when being kidnapped[citation needed]. The comic incorrectly shows Ponytail being allowed to present the results of this study at a conference; reputable scientific journals and conferences should not legitimize studies that clearly violate their ethical norms, such as by failing to obtain informed consent from human subjects before experimenting on them. It is not clear how many people who did agree to participate may have attempted to call the police for assistance regardless; compare the Stanford Prison Experiment.

The title text says that people who didn't agree to go inside an MRI machine are more likely to escape the machine mid-scan. Again, this makes sense, because being put somewhere against your will makes you unsure what is going on, making you try to find a way out.

Transcript

[Ponytail stands on a podium giving a presentation in front of a chart with some box plots.]
Ponytail: Our research shows that compared to the overall population, people who agree to participate in scientific studies are significantly less likely to call the police to rescue them from our lab.


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

Discussion

The selection effect is a bias in the results of a study because the study participants are not a random sample of the general population. For example, a study performed on college students may be biased toward better-educated people, or a study on social interaction may be affected by how many participants have the same first language as the investigators.

The Dining Logician (talk) 19:17, 28 May 2018 (UTC)

One more comic until we hit 2000! Which means on July 11th, the comic number will finally match up with the date (and will certainly be the only time ever). Like an eclipse! Some Commenter (talk) 12:23, 28 May 2018 (UTC)

Just 49 to go until a big round-number milestone! (and just 1 until a big round number kilometerstone) --172.68.54.4 21:54, 28 May 2018 (UTC)
Well, arguably only two more to go, as there was no comic for #404 (Found this out on accident a few minutes ago looking for something unrelated.) But that means the 1000th comic was the 999th too.Linker (talk) 12:47, 29 May 2018 (UTC)

There was no comic #404 because Randall intended the "Page Not Found" error as comic #404. The Dining Logician (talk) 20:18, 29 May 2018 (UTC)

As I recall Mr Munroe's 404th comic was something that he created to mess with us: I remember being surprised by it, looking at the source code of the page and deciding that it was not an error. SDT 108.162.216.52 02:25, 30 May 2018 (UTC) addendum: a /very/ good April fool's joke: Even if it wasn't (it was) he's still got me ;p SDT 108.162.216.52 02:39, 30 May 2018 (UTC)

I don't think escaping from MRI is that easy. There is reason why it's known to be problem for claustrophobic people. -- Hkmaly (talk) 22:46, 28 May 2018 (UTC)

It it not that it is hard to escape from an MRI, unless you are somehow restrained or disabled. It is just that it feels that way because your head is in a tunnel. 141.101.88.220 13:31, 29 May 2018 (UTC)
Is it dangerous to leave an MRI mid-scan? 172.68.189.115 18:57, 29 May 2018 (UTC)
No. Smurfix (talk) 20:42, 30 May 2018 (UTC)

It could be if your escape attempt brings something metallic into the MRI field. The Dining Logician (talk) 19:37, 29 May 2018 (UTC)

Should we mention trivia of comics using the same platform? https://xkcd.com/1781/ 172.68.65.252 16:58, 29 May 2018 (UTC)

https://xkcd.com/1453/ is a simliar, meta-research thing comic. There seems to be several of these now, perhaps a category or at least a mention should be appropriate? 172.69.186.28 08:03, 30 May 2018 (UTC)