Polling Data

I like polls. I think they provide me with information. I think the problem with

No votes yet

A problem with internet polls is they aren't normally considered scientific for many reasons. The biggest is because their "randomness" is questionable at best.

Most polls are on sites that require the Pollee to find it. Thus the only ones to vote are those who either have found or looked for the poll.

Another is because when asking about certain issues the data is skewed because internet users have access to the internet and computers which generally means the poorest of the population are left out. That means the results are automatically skewed as well and thus not a direct reflection of the general public.

The most scientific pollsters will publish their sampling formula to enforce the credibility of their poll. Even through this data can be skewed by the day of the week, areas included, and other factors.

Polls are not considered good science anyway. Too many factors can skew the results. Instead they can be used as good talking points and not much else. Hence why we use voting to determine elections and not just a mass popular vote poll.

MikeyA

MikeyA

Here are two examples of a polls lack of credibility.

Polls shows that people who eat oatmeal daily are more apt to have cancer.

This caused many to falsely believe that oatmeal causes cancer. Instead it just showed a correlation and that's it. The portion of the population that eats oatmeal is normally older. Also the portion of the population that gets cancer is normally older. The result is a correlation between cancer and oatmeal but not a cause effect.

I heard a few guys on the radio discuss this recently. A poll came out which said "Women who wear high heels are more apt to have urinary tract infections" which caused doctors to start swearing to patients not to wear high heels.

The DJ's looked at this from correlation and not cause/effect by asking "Who wears high heels?"

Well normally professional working women wear high heels. This doesn't show anything. But when you add in strippers and prostitutes wear high heels a lot one must wonder about the sample. Strippers and prostitutes normally aren't the most "healthy" people. Thus one could reasonably deduce that they would be more apt to urinary tract infections. So due to the tainted sample population we must question the poll and ask if it's really cause/effect or merely showing a correlation.

MikeyA

MikeyA

that I've experienced with several of your polls, and I made mention of it in the comments section, is that there is rarely an answer that I would have chosen, so I didn't.

I've also noticed some similar comments by others.

Perhaps some better choices would increase the number of responses. Think in broader responses, in both directions.

My 2 cents, FWIW ;-)

Hooda Thunkit

Hooda Thunkit, I suppose I use the poll as a starting point. I may not always have the

Old South End Broadway

When it comes to national polls you definitely have to look at the pollster and their polling philosophy. Those I trust the most are Gallop and Rasmussen. They both regularly post their polling data, past data, and track any changes. Many of the others I find are sketchy and are barely held for more than talking points.

MikeyA

MikeyA

Visited BuzzDash.com, and found it interesting.

MikeyA, I like the national polls, too. But I do like what we are doing here, just wish there were larger numbers so we could get a "snapshot" of what people here think. But we get a lot of that from the comments.

What I might do in future (as Hooda Thunkit pointed out) is submit a "story" about a proposed poll, and ask for suggestions for choices. I don't know if it is laziness, or what, but even though I know I can have more choices than are displayed I usually try to define my poll around the 5 choices that are displayed to be filled out. And, as someone pointed out, often a good poll is just a "yes/no" choice.

Old South End Broadway

I visit BuzzDash.com every once in a while. It's all user submitted polls, to the best of my knowledge.

"good poll is just a "yes/no" choice." Normally but also be careful of wording which can also skew a poll.

Are you in favor of the brutal destruction of a newly formed human life? Yes/No

Do you feel that a woman should be the decision making when it comes to medical choices regarding her body? Yes/No

Both are yes/no questions but both could come up with skewed results due to skewed wording.

MikeyA

MikeyA

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.