• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Poll: Internet Polls

rolleye.gif
 
Rule #1:

Provide SHORT options for your polls, or no one will vote. Alternatively, they will just go for the shortest and easiest to understand option (which is why I AM BATMAN!)

🙂
 
Define "useful": A poll on, "Does my cat's breath smell like catfood?" is hardly useful. So I would say the majority of internet polls, whether or not they are accurate, are not useful.
 
Internet polls are very useful if done properly. Take Harris Poll Online for example. They go to painstaking efforts to make sure they get a random sample of people without any known bias.

Now very few places do that amount of effort. Take political polls at a place like cnn.com for example. I read once that 65% of online poll answers are given by Republicans, the other 35% were Democrats or other. Obviously no matter what, the result will be highly skewed since the poll was based on a skewed population sample. Thus the polls that don't take the effort to remove the bias in the people polled are useless.
 
Internet polls are very useful if done properly...Obviously no matter what, the result will be highly skewed and useless.
Click to expand...
Very...umm...contradictory!

There is no possible way to "go to painstaking efforts to make sure they get a random sample of people without any known bias" on an anonymous internet poll.
 
Very...umm...contradictory!
There is no possible way to "go to painstaking efforts to make sure they get a random sample of people without any known bias" on an anonymous internet poll.
Click to expand...
Ok I edited it to make it clear that those two sentences were talking about different polls. (I thought it was quite clear to begin with, but oh well).

Did you even look at Harris Poll Online? They randomly email online polls to users. But the pool of users that they randomly send the poll to has already been selected to remove bias from income, age, gender, race, location, internet and computer skills, and a ton of other factors. After taking all that into account they have a non-biased pool of people to which they randomly send the poll to a few thousand. That way their polls are highly accurate.

Note: they never once ask your name, social security number, exact address, or anything else that can specifically identify you (other than your email address). So it is very anonymous.

Harris Poll's statement on their statistical analysis:
"We ask questions about your demographics, interests, employment, purchasing behaviors, and online usage to help us develop general trends about our panel. This information is used to conduct statistical analyses so we can ensure that our sample of respondents is representative of the general population. We also use this information to subdivide the panel into special interest groups and to send you surveys that are more pertinent to your interests and characteristics."

I am among the people who take the Harris polls. I like to do it since it gives useful information to companies, they can then better tailor their products to consumer's needs, and for a few minutes a week I get free gift certificates to restaurants I like every couple of months (you can choose other rewards as well).
 
They randomly email online polls to users. But the pool of users that they randomly send the poll to has already been selected to remove bias from income, age, gender, race, location, internet and computer skills, and a ton of other factors. After taking all that into account they have a non-biased pool of people to which they randomly send the poll to a few thousand. That way their polls are highly accurate.
Click to expand...
Ah, well that isn't the kind of poll I was thinking of. I was thinking of those CNN polls on their mainpage where they have no clue about the demographics of the person voting, or how many times that person is voting (clear your cookies), and polls like them.

 
Originally posted by: VBboy
Rule #1:

Provide SHORT options for your polls, or no one will vote. Alternatively, they will just go for the shortest and easiest to understand option (which is why I AM BATMAN!)

🙂
Click to expand...

 
At the risk of bucking the trend I will approach this seriously. If one conducted a perfectly good poll, carefully constructed, there still be no way to guarantee a representative sample participates in it.

Of course this assumes the person sponsoring is trying to mimize bias. Often a "conservative think tank" (or a liberal one) will craft a poll in such a way that it "proves" their position.

Lies, damned lies, and statistics.

PS, there is an old, but still good book called "How to Lie with Statistics" Highly recommended.
 
I started a Poll.... I'm going to analyze the hell out of it and compare it to real statistics.... and tell you guys the correlation!! MUHAHA😀

LOOK DOWN BELOW
\/ \/ \/ \/ \/ \/ \/ \/
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top