PDA

View Full Version : Deception in Science: Good Ethics or Bad Ethics?


JediMaster12
03-18-2008, 12:27 PM
In any research experiment or study, in any discipline, there is a code of ethics that accompanies it. Ethics are used in order to perserve the validity of the study that is being conducted. One of these that is debated among researchers is the notion of deception.

Deception is one of those shades of grey topics in studies and experiementation. More often when it involves people the question is always how much to you reveal to your subjects of participants? On a sidenote: any study that involves people, consent must be obtained. There is such a thing as informed consent that may give the bare bones about what they will be doing and of course assurances of confidentiality are done but how much should a researcher reveal? There is this notion that, if you come from the positivist paradigm (hard science, statistics, causal relationships, etc) if you reveal too much, you can jeoporadize the experiment since people tend to behave differently depending on how much you tell them. There is the idea that they will try to respond in the manner they think you find correct thus not yielding accurate results.

On that note, here is a hypothetical situation: what if a researcher or researchers wanted to study the conversational patterns and online identity perceptions of people on a forum board like Kavar's Corner. They start up an account and create the username and the profile etc. In the course of their interaction they deliberately word their posts to reflect a certain behavior. Would you consider that a violation of ethics?

SilentScope001
03-18-2008, 12:44 PM
I think of it as a good idea to use deception for the greater good. Gathering knowledge should be foremost, and the more utility is generated, the better. The bad news is that this idea would just be abused to justify pretty horrible stuff, and the rest of the scientific community disagrees with the idea nayway, so you might as well obey ethics, if only because you don't want bad PR to occur. Just stay within the gray area, convince the ethics commitee that approves all experiments, and you'll be fine.

HOWEVER, your example would easily be seen as an example of 'trolling' (posting a message to receive a response), and trolling for scientific purposes is still trolling. Not only will you get banned, but people will not treat your results seriously at all. I mean, it's only one sample size of a community that already exist.

No, you want to study communities, you need to pay volunteers to create a brand new internet forum and monitor their behavior.

Jvstice
03-18-2008, 02:01 PM
I do see it as unethical, but I see some of the things that has been done for to have had a payoff that they were seeking that gives a little more leeway, because they couldn't have gotten the same information in any other way.

The whole science of social psychology and it's emergence in the 1960s and 1970's is a prime example. Solomon Ashe's experiment, The Milgram Experiment, the Zimbardo experiment shed a lot of light on human nature and people's willingness to conform in light of things they knew were wrong. I do think that once the foundations of the science was laid, it would be wrong to permit this kind of experiment to continue to be done, ever, because 1) it's deceptive, rather cruel, and upsets people; and 2) it discredits science, the scientific method, and it's followers/practitioners in the minds of much of the public who don't necesarily have any loyalty to the goals that they are trying to achieve.

Jae Onasi
03-18-2008, 04:12 PM
The 'scientist' who posts in a certain way in order to achieve a desired result could be guilty of manipulating subjects to get the answer s/he wants. Studying a group without their consent is likely falling into the 'not ethical' zone, unless you could strip away private data somehow. Evev then it's iffy. Furthermore, if someone tried to do some kind of internet (or other) experiment on my kids without my express consent, I'd be one very unhappy parent. The rules for studying children are more strict, and on a forum like this, there are a lot of mnors.

It's one thing to data-mine from public posts (and it still might require the site owner's permission). It's very different to join a forum with the express purpose of posting in a specific way to test behavior of those forum members without their consent.

Ray Jones
03-18-2008, 07:18 PM
I don't think it is unethical to study the behaviour of the human species in internet forums. I mean I am not posting stuff here to keep it secret. Everyone, scientist or not, can read it and I am aware of that fact. Every day I am judged by my behaviour, at work, at home, in the supermarket. The internet makes no difference. Any imaginary secret experiment makes no difference.
The point is, if *I* behave like an idiot, no matter when or where, it is *never* another persons fault. As long as it does not impose any immoral action I'm fine with playing whatever role in whatever scientific experiment.

Web Rider
03-18-2008, 08:04 PM
It depends on what they're deceiving or who they're deceiving in order to do whatever they're trying to do. If it puts people in danger or it releases private information that should be kept well, private, then it's unethical.

But I don't think there's anything wrong with studying "people" or "things" without being clear and upfront that you're studying them, that influences the study and screws things up.

Jae Onasi
03-18-2008, 08:12 PM
Actually, JediMaster12 didn't specify that the experiment would involve trolling. It could be as simple as 'I want to say something positive about Obama because I bet it'll make the conservatives twitch' or 'I'm going to mention hentai because I bet males age 12-99 are going to drool over pixel bewbs'.

Again, I don't take issue with someone taking public information that's already posted and gleaning information from that. Advertising companies do it all the time. However, if someone is going to intentionally try to manipulate me (or my kids), I want the option to give or deny consent.

SilentScope001
03-18-2008, 08:28 PM
It could be as simple as 'I want to say something positive about Obama because I bet it'll make the conservatives twitch' or 'I'm going to mention hentai because I bet males age 12-99 are going to drool over pixel bewbs'.

Both of which happen to be examples of trolling. The latter is a nontraditional method of trolling, and may be seen as a 'fluff troll' (an example would be posing as a girl to see men droll), but it is still trolling, based on the idea that people are predictable, and that you want to see their reaction.

Trolling: Baiting someone into giving you a desired response.

Jae Onasi
03-18-2008, 08:47 PM
I view trolling as baiting someone into a negative response or making negative comments with the intent of inciting a flame war. Manipulating a post for a desired response doesn't necessarily fall into a negative behavior or trying to elicit a negative behavior.

Jvstice
03-19-2008, 12:21 AM
Actually, a board I currently post on from time to time was founded this way. There is an author that has an official fansite. The author did not keep up with a lot of the specifics of her characters, and spun off a bunch of people who used to post on her fan sites into 2 - 3 anti - fan sites and snark boards.

One of these boards claimed to be from a young man who was engaged to be married from New Orleans. In that time, they decided they wanted to broaden their appeal and move on from being primarily a snark board, and broaden their appeal beyond just making fun of how this one author's character's are all "pod people" and personality wise bear no resemblance to the same character with the same name 10 books earlier within the same series. Instead they became a reading site, of similar genre, but better written novels than the original author the board was organized around.

The board grew and a lot of people came and went. The founder of the board left twice trying to start other boards, then came back, before leaving again for the last time about a year ago. Aparently about 5 years after the first time the board was founded (mid last year), the mods and other administrators for the board discovered that the original founder of the board was a female sociology student from Arkansas by checking the IP address and email address that they'd been in touch with all along.

So this does actually happen.

And yeah. I'd say it's unethical.

JediMaster12
03-19-2008, 01:11 PM
I have always viewed ethics in a rather simple manner. I narrowed it down to don't lie and make sure no harm comes to your subjects. Unfortunately deception is a shade of grey since the amount of information you disclose could affect your results. A professor once labeled the ethics in a seminar as a judgment call. As much as I want to go into research, this has me in a spin. I guess it is as much a judgment call as my professor said.

machievelli
03-19-2008, 02:50 PM
Deception is one of those shades of grey topics in studies and experiementation. More often when it involves people the question is always how much to you reveal to your subjects of participants? On a sidenote: any study that involves people, consent must be obtained. There is such a thing as informed consent that may give the bare bones about what they will be doing and of course assurances of confidentiality are done but how much should a researcher reveal? There is this notion that, if you come from the positivist paradigm (hard science, statistics, causal relationships, etc) if you reveal too much, you can jeoporadize the experiment since people tend to behave differently depending on how much you tell them. There is the idea that they will try to respond in the manner they think you find correct thus not yielding accurate results.

It is a proven fact in sociology and anthropology that knowing you are being observed, your mannerisms are changed by that knowledge. If you don’t believe it, think of all those ‘hidden camera’ shows they use. Would the thief/child abuser/drug dealer actually do what they do if they ‘know’ they are under the camera’s eye?

As a fictional example, Star Trek TNG had an observation post on a planet and they were discovered because of an accident. Once they had dealt with other matters (A local thinking they are the Gods and demanding the return of his dead wife) they showed them the entire structure, then told them they would leave.

SS001 is right and wrong. Unethical behavior is used for a lot of horrible things. The discovery of the Wind Chill is a perfect example. If you don’t know, ask. But since every comment you make can be defined as trolling by someone who wishes to make it so.

However:

I think of it as a good idea to use deception for the greater good.

Is the same excuse used by filmmaker Michael Moore when challenged over the ‘facts’ used in Bowling for Columbine and Fahrenheit 9-11.

After all, it is called political 'science'

SilentScope001
03-19-2008, 03:34 PM
After all, it is called political 'science'

Political Science isn't really that 'scientific'.

Besides, deception for the use of a scientific experiment is far different than deception for the use of getting people to believe in your ideology. The former is good because you are trying to find out the truth about society. The latter is trying to make others believe in your ideology and you are not interested in trying to figure out how society runs (because you already believe you know the truth, you just want to persuade others to it).

The former is okay. The latter...er...not so much.

Jae Onasi
03-19-2008, 04:23 PM
When you're talking about doing experiments on human subjects, however, you do need to get informed consent. If you want to do forum experiments, all you have to do is put in the rules "This forum is being monitored as part of ongoing research being conducted by the University of X. By posting on this forum, you agree that information in your posts may be used in our studies. No private information will be released."

It's very simple, doesn't give away the nature of the study, and allows people to opt out if they don't want to be a part of a study.

Ray Jones
03-19-2008, 04:50 PM
If you don't want private information to be released anywhere, don't post about it in public internet forums in the first place.

Inyri
03-19-2008, 05:00 PM
If you don't want private information to be released anywhere, don't post about it in public internet forums in the first place.And don't have credit cards, don't have a license, don't go to school, don't go to the grocery stores... Come on, be realistic.

Ray Jones
03-19-2008, 05:11 PM
Hey, I am! :) There is a difference between making private information and data puclic in an internet forum and and giving these to a financial institute.