1390: "Research Ethics"

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

Moderators: Moderators General, Prelates, Magistrates

User avatar
Freiberg
Posts: 171
Joined: Thu Oct 30, 2008 4:09 am UTC

1390: "Research Ethics"

Postby Freiberg » Fri Jul 04, 2014 4:29 am UTC

Image

Title text: "I mean, it's not like we could just demand to see the code that's governing our lives. What right do we have to poke around in Facebook's private affairs like that?"

At first, I thought that everybody was overreacting to this: social media companies test things out all the time, and in any event this only caused 0.1% shifts in public moods, only detectable with their huge sample size. Then I realized that while large manipulation is obvious, almost nobody would even notice such subtle changes like that.

Now I don't know what to think.

teelo
Posts: 781
Joined: Thu Apr 08, 2010 11:50 pm UTC

Re: 1390: "Research Ethics"

Postby teelo » Fri Jul 04, 2014 4:33 am UTC

In before Randall fixes the double "what"s.

User avatar
Freiberg
Posts: 171
Joined: Thu Oct 30, 2008 4:09 am UTC

Re: 1390: "Research Ethics"

Postby Freiberg » Fri Jul 04, 2014 4:39 am UTC

Will he fix that? I thought that was just his own subtle experiment...

User avatar
Envelope Generator
Posts: 582
Joined: Sat Mar 03, 2012 8:07 am UTC
Location: pareidolia

Re: 1390: "Research Ethics"

Postby Envelope Generator » Fri Jul 04, 2014 5:00 am UTC

The uncorrected original will be collectible! SAVE AS! Quick, before he notices!
I'm going to step off the LEM now... here we are, Pismo Beach and all the clams we can eat

eSOANEM wrote:If Fonzie's on the order of 100 zeptokelvin, I think he has bigger problems than difracting through doors.

User avatar
BlitzGirl
Posts: 9089
Joined: Mon Sep 20, 2010 11:48 am UTC
Location: Out of the basement for Yip 6! Schizoblitz: 115/2672 NP
Contact:

Re: 1390: "Research Ethics"

Postby BlitzGirl » Fri Jul 04, 2014 5:02 am UTC

*puts on British accent*

Looks like we've got some mustard, what what?
Bets on how long before it gets fixed?

Edit for posterity, the current version:

Image
Last edited by BlitzGirl on Fri Jul 04, 2014 2:04 pm UTC, edited 1 time in total.
Knight Temporal of the One True Comic
BlitzGirl the Pink, Mopey Molpy Mome
Spoiler:
Image
Image
Image<Profile
~.Image~.FAQ->Image

User avatar
Envelope Generator
Posts: 582
Joined: Sat Mar 03, 2012 8:07 am UTC
Location: pareidolia

Re: 1390: "Research Ethics"

Postby Envelope Generator » Fri Jul 04, 2014 5:06 am UTC

Wait for for it.
I'm going to step off the LEM now... here we are, Pismo Beach and all the clams we can eat

eSOANEM wrote:If Fonzie's on the order of 100 zeptokelvin, I think he has bigger problems than difracting through doors.

User avatar
BlitzGirl
Posts: 9089
Joined: Mon Sep 20, 2010 11:48 am UTC
Location: Out of the basement for Yip 6! Schizoblitz: 115/2672 NP
Contact:

Re: 1390: "Research Ethics"

Postby BlitzGirl » Fri Jul 04, 2014 5:08 am UTC

Perhaps this should have been titled "Research Ethics Ethics."
Knight Temporal of the One True Comic
BlitzGirl the Pink, Mopey Molpy Mome
Spoiler:
Image
Image
Image<Profile
~.Image~.FAQ->Image

LonePaladin
Posts: 3
Joined: Fri Apr 11, 2014 8:34 am UTC

Re: 1390: "Research Ethics"

Postby LonePaladin » Fri Jul 04, 2014 5:37 am UTC

It's the Double Double Dude Dude Ranch Ranch!

(I've been watching way too much Sesame Street with my son. I'd provide a link, but that makes the system think it's spam.)

User avatar
Klear
Posts: 1965
Joined: Sun Jun 13, 2010 8:43 am UTC
Location: Prague

Re: 1390: "Research Ethics"

Postby Klear » Fri Jul 04, 2014 5:41 am UTC

Damn. I came here mainly to point out the the mistake, but turns out people are talking about nothing else.

Edit: @LonePaladin That happens only to people who forget to read the rules.

nunks
Posts: 3
Joined: Tue Oct 27, 2009 3:53 am UTC

Re: 1390: "Research Ethics"

Postby nunks » Fri Jul 04, 2014 5:56 am UTC

Freiberg wrote:At first, I thought that everybody was overreacting to this: social media companies test things out all the time, and in any event this only caused 0.1% shifts in public moods, only detectable with their huge sample size. Then I realized that while large manipulation is obvious, almost nobody would even notice such subtle changes like that.

Now I don't know what to think.



I think Facebook should answer for the risks this kind of experiment entails. You never know what could be the last straw for someone on the edge, and the larger the scope, the larger the risks of someone killing oneself or someone else. The fact that most changes would stay unnoticed while affecting the discernment of the users is what scares me the most.

Then again, it has always been difficult to know what's true and what's not.

Antior
Posts: 61
Joined: Wed Jan 18, 2012 10:34 pm UTC

Re: 1390: "Research Ethics"

Postby Antior » Fri Jul 04, 2014 6:22 am UTC

The research isn't valid, anyway.

According to my own research, there aren't any sane people on Facebook. So it's not a valid sample of the population.

obfpen
Posts: 59
Joined: Mon Jun 20, 2011 1:16 pm UTC

Re: 1390: "Research Ethics"

Postby obfpen » Fri Jul 04, 2014 6:42 am UTC

No, Randall mustn't be allowed to "fix" the what what "mistake". Then there'd be two groups of readers who see different things (assuming there aren't different groups already; there is precedent here) and who knows what nefarious things will be done with the data that generates.

Eutychus
Posts: 446
Joined: Mon Jan 25, 2010 6:01 am UTC
Location: France

Re: 1390: "Research Ethics"

Postby Eutychus » Fri Jul 04, 2014 6:54 am UTC

What what are we talking about?
Be very careful about rectilinear assumptions. Raptors could be hiding there - ucim

User avatar
Red Hal
Magically Delicious
Posts: 1445
Joined: Wed Nov 28, 2007 2:42 pm UTC

Re: 1390: "Research Ethics"

Postby Red Hal » Fri Jul 04, 2014 7:06 am UTC

The title text is the most interesting to me, because the agreement we signed when we joined up with Facebook does let them do things like this. We signed away our privacy, Facebook didn't.
Lost Greatest Silent Baby X Y Z. "There is no one who loves pain itself, who seeks after it and wants to have it, simply because it is pain..."

User avatar
Flumble
Yes Man
Posts: 2229
Joined: Sun Aug 05, 2012 9:35 pm UTC

Re: 1390: "Research Ethics"

Postby Flumble » Fri Jul 04, 2014 7:10 am UTC

Envelope Generator wrote:What for for it.

Fixed what for you. :roll:


I'm in the dark here: what (what) did facebook do this time that it's upsetting some people?

sol_hsa
Posts: 9
Joined: Mon Nov 19, 2012 8:07 am UTC
Location: Nowhere whenever

Re: 1390: "Research Ethics"

Postby sol_hsa » Fri Jul 04, 2014 7:21 am UTC

Eutychus wrote:What what are we talking about?

Mostly the 'what what' word doubling.

But also the recent debacle regarding one paper published by facebook, wherein they manipulated random people's news feeds, and noticed that if you see a lot of negative stuff, you tend to post negative stuff, and vice versa. Here's some slashdot on the subject: http://news.slashdot.org/story/14/07/02 ... and-frenzy

User avatar
Khrushy
Posts: 48
Joined: Mon Feb 25, 2013 7:00 am UTC

Re: 1390: "Research Ethics"

Postby Khrushy » Fri Jul 04, 2014 7:21 am UTC

Antior wrote:The research isn't valid, anyway.

According to my own research, there aren't any sane people on Facebook. So it's not a valid sample of the population.


According to my own research, there are very few sane people in the general population, so I don't that would invalidate the sample.
Madness takes its toll, please have exact change.

User avatar
BlitzGirl
Posts: 9089
Joined: Mon Sep 20, 2010 11:48 am UTC
Location: Out of the basement for Yip 6! Schizoblitz: 115/2672 NP
Contact:

Re: 1390: "Research Ethics"

Postby BlitzGirl » Fri Jul 04, 2014 7:27 am UTC

Khrushy wrote:so I don't that would invalidate the sample.

I think think I stole your verb.
Knight Temporal of the One True Comic
BlitzGirl the Pink, Mopey Molpy Mome
Spoiler:
Image
Image
Image<Profile
~.Image~.FAQ->Image

User avatar
Flumble
Yes Man
Posts: 2229
Joined: Sun Aug 05, 2012 9:35 pm UTC

Re: 1390: "Research Ethics"

Postby Flumble » Fri Jul 04, 2014 7:38 am UTC

BlitzGirl wrote:
Khrushy wrote:so I don't that would invalidate the sample.

I think think I stole your verb.

You cheeky hat-person!

Invertin
Posts: 127
Joined: Fri Mar 20, 2009 8:08 pm UTC

Re: 1390: "Research Ethics"

Postby Invertin » Fri Jul 04, 2014 8:01 am UTC

I don't know the law on such things in america but I was taught that a sociological/psychological test is automatically considered 'unethical' unless people are told what it entails and are allowed to decide whether or not their data is used in the actual results.

I don't remember the exact definition of ethical vs unethical by legal standards but at the very least that is one of them.

sol_hsa
Posts: 9
Joined: Mon Nov 19, 2012 8:07 am UTC
Location: Nowhere whenever

Re: 1390: "Research Ethics"

Postby sol_hsa » Fri Jul 04, 2014 8:13 am UTC

Invertin wrote:I don't remember the exact definition of ethical vs unethical by legal standards but at the very least that is one of them.


Ah.. remember the golden rule: who has the gold, makes the rules. ;)

User avatar
orthogon
Posts: 3059
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1390: "Research Ethics"

Postby orthogon » Fri Jul 04, 2014 8:23 am UTC

Two observations:
1. The first "what" looks as though it was added later - it's smaller and, perhaps, strangely kemed.
2. I found the first sentence quite hard to parse. I thought Facebook was showing us [how] to conduct the research. Was it just me?
xtifr wrote:... and orthogon merely sounds undecided.

speising
Posts: 2349
Joined: Mon Sep 03, 2012 4:54 pm UTC
Location: wien

Re: 1390: "Research Ethics"

Postby speising » Fri Jul 04, 2014 8:28 am UTC

randall, of course, is doing similar manipulations regularly with us. there's no other explanation for the AGW-denier baiting, for example.

User avatar
Klear
Posts: 1965
Joined: Sun Jun 13, 2010 8:43 am UTC
Location: Prague

Re: 1390: "Research Ethics"

Postby Klear » Fri Jul 04, 2014 8:39 am UTC

Antior wrote:The research isn't valid, anyway.

According to my own research, there aren't any sane people on Facebook. So it's not a valid sample of the population.


But a huge amount of people are on facebook. Are you saying that people are insane?

orthogon wrote:2. I found the first sentence quite hard to parse. I thought Facebook was showing us [how] to conduct the research. Was it just me?


Not just you. I wanted to say the same thing, but I waited so I wouldn't have to double edit.

User avatar
HES
Posts: 4888
Joined: Fri May 10, 2013 7:13 pm UTC
Location: England

Re: 1390: "Research Ethics"

Postby HES » Fri Jul 04, 2014 10:13 am UTC

obfpen wrote:No, Randall mustn't be allowed to "fix" the what what "mistake". Then there'd be two groups of readers who see different things (assuming there aren't different groups already; there is precedent here) and who knows what nefarious things will be done with the data that generates.

Not that he's ever done that before.
He/Him/His Image

User avatar
Neil_Boekend
Posts: 3220
Joined: Fri Mar 01, 2013 6:35 am UTC
Location: Yes.

Re: 1390: "Research Ethics"

Postby Neil_Boekend » Fri Jul 04, 2014 10:32 am UTC

Klear wrote:But a huge amount of people are on facebook. Are you saying that people are insane?

I do hope so. Else I'd be left out.
Mikeski wrote:A "What If" update is never late. Nor is it early. It is posted precisely when it should be.

patzer's signature wrote:
flicky1991 wrote:I'm being quoted too much!

he/him/his

Morgan Wick
Posts: 39
Joined: Fri Feb 15, 2013 9:21 pm UTC

Re: 1390: "Research Ethics"

Postby Morgan Wick » Fri Jul 04, 2014 10:38 am UTC

Comments from this point and below are from after the what what error was corrected.

User avatar
Flumble
Yes Man
Posts: 2229
Joined: Sun Aug 05, 2012 9:35 pm UTC

Re: 1390: "Research Ethics"

Postby Flumble » Fri Jul 04, 2014 11:05 am UTC

What what? He actually did revise the comic just now? Again?

teelo
Posts: 781
Joined: Thu Apr 08, 2010 11:50 pm UTC

Re: 1390: "Research Ethics"

Postby teelo » Fri Jul 04, 2014 11:18 am UTC

teelo wrote:In before Randall fixes the double "what"s.

Called it.

teelo
Posts: 781
Joined: Thu Apr 08, 2010 11:50 pm UTC

Re: 1390: "Research Ethics"

Postby teelo » Fri Jul 04, 2014 11:19 am UTC

Flumble wrote:What what? He actually did revise the comic just now? Again?

I posted the old image in this thread: viewtopic.php?f=2&t=22741&start=7640

obfpen
Posts: 59
Joined: Mon Jun 20, 2011 1:16 pm UTC

Re: 1390: "Research Ethics"

Postby obfpen » Fri Jul 04, 2014 11:43 am UTC

I just saw the comic with the second what removed, and now I feel slightly worse. Has anyone seen the first what removed? If so, how did that affect your emotional state?*

* By answering this question you agree that the data you provide may be used by nefarious porpoises conducting research into the tolerance of land-based bipedal mammals to minor linguistic manipulations.
Last edited by obfpen on Fri Jul 04, 2014 11:48 am UTC, edited 1 time in total.

User avatar
Klear
Posts: 1965
Joined: Sun Jun 13, 2010 8:43 am UTC
Location: Prague

Re: 1390: "Research Ethics"

Postby Klear » Fri Jul 04, 2014 11:44 am UTC

For those of you who didn't see the comic before it was first fixed, here it is:

Image

Dr. Gamera
Posts: 20
Joined: Wed Mar 28, 2012 2:23 pm UTC

Re: 1390: "Research Ethics"

Postby Dr. Gamera » Fri Jul 04, 2014 1:38 pm UTC

Facebook lost me when they made it difficult, if not impossible, to simply see all my friends' status updates in chronological order, most recent first. Please don't guess at what I want to see most; I wouldn't even want another human being doing it, much less an algorithm in this age of weak AI.

User avatar
HES
Posts: 4888
Joined: Fri May 10, 2013 7:13 pm UTC
Location: England

Re: 1390: "Research Ethics"

Postby HES » Fri Jul 04, 2014 1:43 pm UTC

Dr. Gamera wrote:Facebook lost me when they made it difficult, if not impossible, to simply see all my friends' status updates in chronological order, most recent first. Please don't guess at what I want to see most; I wouldn't even want another human being doing it, much less an algorithm in this age of weak AI.

This is by far their most annoying development. I've persuaded my computer to stick to "most recent", but have to set it manually each time on my phone. Even then I'm not convinced I'm being shown everything.
He/Him/His Image

User avatar
orthogon
Posts: 3059
Joined: Thu May 17, 2012 7:52 am UTC
Location: The Airy 1830 ellipsoid

Re: 1390: "Research Ethics"

Postby orthogon » Fri Jul 04, 2014 1:55 pm UTC

Klear wrote:For those of you who didn't see the comic before it was first fixed, here it is:

Spoiler:
Image


:) That's how most xkcds appear to the average non-geek...
xtifr wrote:... and orthogon merely sounds undecided.

User avatar
PinkShinyRose
Posts: 834
Joined: Mon Nov 05, 2012 6:54 pm UTC
Location: the Netherlands

Re: 1390: "Research Ethics"

Postby PinkShinyRose » Fri Jul 04, 2014 3:15 pm UTC

The article doesn't mention review by an ethics committee. Then again, I get the idea that psychology ethics committees tend to approve anything so it probably wouldn't have made a difference.

gimmespamnow
Posts: 49
Joined: Wed Sep 14, 2011 6:35 am UTC

Re: 1390: "Research Ethics"

Postby gimmespamnow » Fri Jul 04, 2014 5:54 pm UTC

Facebook conducts experiments all the time on its user base, this isn't new. Most of the experiments have to do with ad placement or sponsored content and so on, basically, they are trying to figure out the best way to manipulate users to make more money. I don't know of any medium to large company that doesn't sometime try an experiment on customers and measure the results, and sure some of the experiments are pretty boring, (what time and day of the week is good to send out a marketing newsletter?) but all those things are ultimately psychological experiments...

What makes this Facebook experiment special is that is doesn't seem to have been conducted by the marketing department and doesn't have direct marketing value. You can say "this manipulated people's emotions" but most of the spam I get is designed to make me feel like I'm small in the manhood department or make me think that I can get rich by just giving someone my bank account info etc, are you going to claim that those AREN'T designed to manipulate me?

User avatar
cellocgw
Posts: 2044
Joined: Sat Jun 21, 2008 7:40 pm UTC

Re: 1390: "Research Ethics"

Postby cellocgw » Fri Jul 04, 2014 6:22 pm UTC

Eutychus wrote:What what are we talking about?


No, it's "Which what are we talking about?" :roll:


ETA:
1) Gosh a lot of OTTers popped up on this thread,

2) So you really think GLR isn't doing massive emotion-modification via his comics feedback software? :twisted:
Last edited by cellocgw on Fri Jul 04, 2014 6:25 pm UTC, edited 1 time in total.
https://app.box.com/witthoftresume
Former OTTer
Vote cellocgw for President 2020. #ScienceintheWhiteHouse http://cellocgw.wordpress.com
"The Planck length is 3.81779e-33 picas." -- keithl
" Earth weighs almost exactly π milliJupiters" -- what-if #146, note 7

User avatar
cellocgw
Posts: 2044
Joined: Sat Jun 21, 2008 7:40 pm UTC

Re: 1390: "Research Ethics"

Postby cellocgw » Fri Jul 04, 2014 6:24 pm UTC

Klear wrote:For those of you who didn't see the comic before it was first fixed, here it is:

Image


You left out a ",Ginger" at the end.

(for those whom that mystifies, track down "blah blah Ginger" in the Far Side archives)
https://app.box.com/witthoftresume
Former OTTer
Vote cellocgw for President 2020. #ScienceintheWhiteHouse http://cellocgw.wordpress.com
"The Planck length is 3.81779e-33 picas." -- keithl
" Earth weighs almost exactly π milliJupiters" -- what-if #146, note 7

User avatar
selene
Posts: 14
Joined: Fri Jun 06, 2008 4:27 am UTC

Re: 1390: "Research Ethics"

Postby selene » Fri Jul 04, 2014 7:09 pm UTC

Whether Facebook's actions were ethical or not, they were stupid. Facebook manipulated users in secret and then announced it. As I understand the story, individual users don't even know whether they were subjects of the study or not. Which makes every single facebook user feel potentially violated.

Companies manipulate our emotions all the time, from advertisements to product names to the color of a label. We expect this. What Facebook did that is so unusual is publish the results as academic research instead of (or in addition to?) using them to convince us to spend more money.

Academic studies that use human subjects are required to obtain informed consent. If informed consent cannot be given - such as when the subjects are young children, or psych studies where awareness would change the results - then the researchers need to get approval from an Institutional Review Board.
These rules do not apply to private companies. The only judge of their ethics is their customers. When this furor dies down we'll see whether anyone has lost trust in Facebook. (To judge from the reactions so far to this comic, no one actually cares what Facebook did).

What Facebook should have done is treat the subjects of its academic study like academic subjects. Get approval from the IRB of one of the universities that advised on the study. Get informed consent to use the data after collecting it. Maybe even compensate the subjects (with extra photo space or something).


Return to “Individual XKCD Comic Threads”

Who is online

Users browsing this forum: chridd and 27 guests