DU this Freep

November 18, 2005 11:04 AM

Help me understand some behavior exhibited in online politics forums.

Participation on sites such as Democratic Underground and Free Republic often seems to involve trying to skew online polls in the direction of their political views. I don't really understand the benefit of doing this.

Wikipedia says "Some forum posts are aimed at influencing polls on other websites. Media websites (including newspapers, television networks, and America Online), run occasional "polls" that do not use the sampling methods of formal opinion polls, but instead invite everyone to respond. Some DU forum messages, usually captioned "DU this poll", urge DU members to vote en masse in such an online poll. The practice of trying to affect the poll results is not unique to the Democratic Underground forums and is employed by other activist websites of all political stripes."

This still doesn't answer my question... What are they hoping to accomplish? Do members of either site really think that policy decisions or major reporting hinge upon the results of an unscientific poll?

posted by trey to Computers & Internet (16 answers total)

They're not trying to influence policy directly, they're trying to influence more opinion. Many people are followers. They form opinions based on how other people feel, or feel justified in their own opinions if popular support seems to back them up. By skewing poll results that get reported in the media, you create the illusion of agreement with your opinions.
posted by jacquilynne at 11:07 AM on November 18, 2005

Most Americans that vote tend to vote for the most popular thing. The most popular thing is the thing that wins the polls. Bandwagons are fun!
posted by cleverusername at 11:07 AM on November 18, 2005

Most people do not know how to think for themselves and therefore expect to be told what to think. Unfortunately, political activists and the religious right are taking advantage of this.

I built WatchBlog.com in 2003 as a place for people to debate politics in a civil and intelligent manner, and to get people to understand that there are other political perspectives than the one they think they believe in. Trolls and narrow-minded thinkers are asked to leave by enforcing a strict "critique the message, not the messenger" policy.
posted by camworld at 11:37 AM on November 18, 2005

I seem to remember several news stories that report the results of polls, in the form of "an online poll said that 88% of Americans eat boogers..." and then at the end of the piece "We should remind viewers that this isn't a scientific poll."

Ah yes, my local TV news is guilty of this and so is the BBC.

If Voltaire were alive today he might have said "A witty online poll proves nothing."
posted by revgeorge at 11:47 AM on November 18, 2005

Most people do not know how to think for themselves and therefore expect to be told what to think. Unfortunately, political activists and the religious right are taking advantage of this.

I guess you've never read Democratic Underground. It is the very definition of groupthink.

-
posted by Independent Scholarship at 11:50 AM on November 18, 2005

This behavior is not unique to political sites. Fan forums, entertainment sites, all sorts of online communities tied to a specific person or entity or philosophy engage in this meaningless poll-stuffing. A webcomic up for a "reader's choice" award may well exort it's readers, or see an unsolicited swell of reader organization, to get the vote out early and often, for example.

And while you're right that a cnn online poll or such lacks even the award-giving feedback of the webcomic example, the principle is the same -- they have this horse, and this horse is in a race, and so they're invested in the race. It's an easy thing to do and validates however symbolically their preferences.

Plus people have a lot of time on their hands. Plus people actually enjoy gaming things just for the sake of gaming them. And the ability to organize your site-members to game a poll in the direction of your choosing reinforces a sense of righteous community even when the pointlessness of the exercise from a policy perspective is known.
posted by cortex at 11:52 AM on November 18, 2005

People are naturally inclined to conform to the social norm, even when it conflicts with their own best judgment. For example, a group of individuals are asked to identify which of three lines is the same length as a reference line. If performed alone, most are able to give the correct answer. However if they first hear their peers give the incorrect answer, they frequently give the answer their peers gave.

There probably are, or were, evolutionary advantages to "group think".

I suspect everyone has succumb to its effect at points in their lives.
posted by justkevin at 11:52 AM on November 18, 2005

Skewing polls comes from the public impression that argumentum ad populum is valid.
posted by Dipsomaniac at 2:09 PM on November 18, 2005

(Yeah, there's certainly herd activity on both sides)
posted by johngoren at 2:32 PM on November 18, 2005

This behavior is not unique to political sites. Fan forums, entertainment sites, all sorts of online communities tied to a specific person or entity or philosophy engage in this meaningless poll-stuffing. A webcomic up for a "reader's choice" award may well exort it's readers, or see an unsolicited swell of reader organization, to get the vote out early and often, for example.


You see this a lot if you hang around band websites - people going so far as to delete the particular poll cookie so they can vote as often as they have the patience (in a particular british mag's online poll, they had to disqualify Muse because a handful of fans from a particular site voted over a million times!) (note: We should remind viewers that this isn't a scientific example.)
posted by muddgirl at 3:01 PM on November 18, 2005

I don't think it's strictly a "herd mentality" thing they're hoping for, or just trying to sway non-freepers/DUers opinions; it's more about creating the illusion of a mandate.

What would be more egregious to you if the president was going to enact a new policy that you were opposed to:
1- if he were doing it with wide-spread support
2- if most citizens (& even Poland) were opposed?
Even if I were against the policy, if he had a *true* mandate, to some degree I'd have less of a problem with him enacting it than if he did it without any concern that a sizable chunk of his constituency's opinion indicated he ought to do otherwise.
posted by neda at 3:15 PM on November 18, 2005

Best answer: Do members of either site really think that policy decisions or major reporting hinge upon the results of an unscientific poll?

A lot of it is insulation from a representation of truth. It's not like these people believe the "scientific" polls either. There's also the sense of participating in some kind of special op, like an electronic microprotest, something very small and easy you can do to increase the volume of your point of view.

You could also turn this around and ask why do news organizations and other things bother with these crappy online polls anyway. Do they really make decisions based on them? If they're merely passing them on to the readers, do they really think readers should make decisions based on them? Or are they just mindless entertainment anyway? In that case maybe the question is why would anyone ever vote in or pay attention to an online poll of this kind?

So a glib answer to your original question is "because they're there, and it's possible," but I think the question should be asked why are they there, anyway?
posted by fleacircus at 4:19 PM on November 18, 2005

While online polls are unscientific, they do resemble the way elections work. People may not expect to effect change by voting in a poll, but being able to "turn out the vote" and show the strength of your group in its numbers is what political campaigns do. Political campaigns try to construct a majority of vote-casters on election day, even if that group of voters isn't a "sample" of the population. So, these political groups are perhaps trying to show they have enough power in their numbers to skew a poll, not convince people of an illusory popularity.

Furthermore, they may have no other audience than themselves, as skewing a poll by working in collaboration with others builds solidarity within the group. It's like street gangs, or soldiers in dress uniform puffing their chests out in military parades, or mock warfare.
posted by rschram at 4:33 PM on November 18, 2005

It's possible they're not even thinking as far as "Let's ballot-stuff this poll to influence impressionable people!" more like "Let's ballot-stuff this poll because if our view comes out ahead, that means WE WIN!" People like to see their opinions expressed.
posted by dagnyscott at 6:55 PM on November 18, 2005

So a glib answer to your original question is "because they're there, and it's possible," but I think the question should be asked why are they there, anyway?

Because the first and easier step to involve a reader is to make him/her click on something. And a poll is the easiest click-trap.

Most mass media have no idea what participation of the readers to the content means. They have no idea how user generated content works. Look at the NYTimes, look even at Salon: every traditionnal media knows only how to broadcast his own content and publishes only snippets of content generated by users.

But they read everywhere that on this intraweb thingy they have to make something interactive. Hey, I made a poll, see? My website is interactive, yeah! My readers will now think that I care for their opinion and will let us continue to broadcast our own manufactured content the same old way.

Hence the poll. Click.
posted by bru at 7:02 PM on November 18, 2005

Here's the thing. People like to believe that their opinions are supported by hard evidence rather than prejudice. Statistical evidence can help there, because it carries the impression of meaningful analysis. This impression may or may not be valid in any given case.

Here's the problem: as we all know, statistical evidence can be good, bad or inconclusive ("...lies, damned lies and...").

But. If you're pushing a political (or other) agenda and you feel really strongly about it, it's easy to get into an "ends justify the means" mindset, and statistical studies are one of the easiest means to support your ends. This applies whether the quality of those studies is good or bad. Many people simply do not understand the underlying maths and how easily it can be used to distort reality. Statistical studies are therefore an incredibly useful tool for advancing a cause.
posted by Decani at 7:01 AM on November 19, 2005