Subject:
|
Polls as propaganda devices.
|
Newsgroups:
|
lugnet.admin.general
|
Date:
|
Wed, 14 Aug 2002 04:20:42 GMT
|
Highlighted:
|
(details)
|
Viewed:
|
185 times
|
| |
|
|
Let me preface this by saying that the polling mechanism just introduced,
while it has some bumps (some of which have already been fixed, like the
ability to keep a poll result anonymous, e.g. the CLSotW poll results are
now anonymous so everyone go vote, eh? ), is really really nifty, technically.
But I'm wondering a bit what the point of polls are. Perhaps the point is
"polls are anything you want them to be".
That seems to be the case. We have some polls I would characterise as being
purely silly, or purely for fun (e.g. the bees peas cheese knees poll) and
some are more serious. Some are narrowly focused, and some are quite a bit
broader. There are a few that I think will gather good information for
external parties, and a number that will advance fan projects (e.g. the
Technic hall of fame).
All well and good.
But the point of this post is none of the above. Rather it's to wonder about
the efficacy of using a poll as a propaganda device.
A propaganda device, you say?
It is possible, when constructing a poll, to structure the questions so as
to try to influence the pollee to a certain outcome. That's in some ways
unavoidable, and it is very hard not to let bias show through...
For example, consider http://members.lugnet.com/polls/ballots/?n=3 which is
a very narrow poll. It asks a specific question, and it is hoped that the
result will be of some use to LD as well as to third parties (if someone
else enters the track market for 9V track I am going to be buying lots,
although my expectation is nil) and to fans. My bias nevertheless may have
leaked through... you be the judge.
But I would argue that poll #3 at least covers all the possible answers,
even if bias shows through. (my bias is that there *is* a problem with the
track geometry and it needs fixing, but I admit that others may not see it
that way and have allowed for several gradations including an answer of the
form "everything is fine")
So I would not characterise #3 as a propaganda device, although I could be
wrong. It's my attempt to see how serious the geometry problem is perceived
to be among the self selected sample set that took the poll.
Some people did in fact answer "there is no problem" and that's fine.
Now consider http://members.lugnet.com/polls/ballots/?n=21 ... Look
carefully at the answers. This is a much broader question, of course, but
the answer set has been constructed so that it deliberately or inadvertantly
does not cover all possible responses. There is no "none of the above", no
"TLC's current direction is fine", no "I enjoy video games, bring me more
LEGO themed games", no "juniorization is a good thing" or similar. The
author's bias shows through clearly, all answers support the position that
he has taken in the past. If you answer this poll and choose ANY answer, you
are supporting the author's case. (1) The only way not to do so is not to
answer at all.
Now, there's nothing inherently wrong with that. Or is there?
That brings me back to "what's the point of polls?" If they're just a
mechanism to have some fun, great. But if anyone thinks that they can use
poll results to make any broad case (anything beyond "the pollees like this
set better than that set" sorts of questions, that is) I think they are
greatly mistaken, unless the poll has been constructed with some rigor. (2)
Even so it's a self selected sample set which has its own drawbacks.
So I guess this post is intended to raise awareness... in case you hadn't
figured it out yet, some pollsters here on LUGNET(tm) using this mechanism
have axes of various sizes to grind and are using polls to do just that
rather than to have fun. (3)
Some show some bias but have tried to construct fair surveys anyway to get
around it. Some show lots and lots of bias and have constructed misleading
surveys. Nothing wrong with that if you realise it up front... but forearmed
is half an octopus, as they say.
It's also a notice that if people use these polls as support for arguments,
they can expect to be called on their survey methodology.
1 - In fact I share much of the author's bias and would have chosen "all of
the above" IF a "no, there is no problem" choice had been included, but
since the poll is fatally flawed (4) I have chosen not to take it at all.
Right now we have no way to withdraw a poll so whatever polls are created
are out there for good, unless something is changed in the poll implementation.
2 - Google is always a good place to go. For a possible starting point to
read about polling techniques, I found this reference which I have not
studied in depth but which does align with what I am getting at:
http://writing.colostate.edu/references/research/survey/pop2f.cfm ,
particularly this section
http://writing.colostate.edu/references/research/survey/com4a2c1.cfm
3 - To those that already knew that, my apologies for stating the obvious.
4 - as a poll, not as a propaganda device, it's quite clever as a propaganda
device
PS, Todd... a cite from it about what not to do: "Questions with long
lists-these questions may tire respondents or respondents may lose track of
the question" ... that language survey of yours tired me out! Grin. You were
just testing how many items you can have before you break the mechanism, right?
|
|
Message has 3 Replies:
 | | Re: Polls as propaganda devices.
|
| (...) Well, you could include a standard quote at the bottom, similar to the one at on slashdot: (URL) Don't complain about lack of options. You've got to pick a few when you do multiple choice. Those are the breaks. * Feel free to suggest poll (...) (22 years ago, 14-Aug-02, to lugnet.admin.general)
| | | Re: Polls as propaganda devices.
|
| (...) Actually, I had a bit of trouble with this poll. The reason is that I am now not convinced LEGO should put resources into alternative track geometries. On the other hand, it would be real swift to have some new track geometries, so my personal (...) (22 years ago, 14-Aug-02, to lugnet.admin.general)
|
11 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
