Online Poll Trickery

[OneAdam12]

I just clicked on the Newsmax poll link on O.com Where it asked if Obama should be re-elected. Here are the questions. A little poll question trickery is involved here.

1) Do you support the full repeal of President Obama's healthcare plan Congress passed in 2010?

Yes, repeal it completely
No, don’t repeal it at all
No, repeal just parts of it

2) Should Congress restore $500 billion in Medicare benefits for seniors that the Obama plan cut?

Yes, restore the Medicare benefits
No, don't restore the Medicare benefits

3) Do you support President Obama's desire to give 12 million illegal aliens amnesty and a path to citizenship?

Yes, I support Obama's amnesty plan
No, I oppose it

4) What do you believe Congress' top priority should be in 2011?

Cutting spending
Cutting taxes
Paying down the national debt
Improving the nation's security

5) Do you plan on voting to re-elect President Obama in 2012?

Yes, re-elect
No, for another candidate
Not sure right now

6) Who did you vote for in 2008?

McCain - Palin
Obama - Biden
Other

Read more on Newsmax.com: Should Congress Repeal Obama's Policies
Important: Do You Support Pres. Obama's Re-Election? Vote Here Now!

[phillydog07]

I'm not seeing the trickery, but maybe I'm not looking close enough. The only thing I see is that some question do not have options for "unsure" or "other". I wouldn't call that trickery per se, just poor survey design.

[westside popo]

What ever it is, it's a lousy worded biased poll.

[Bounce]

It "trains" the testee with the leading questions up to the one about voting to re-elect O. Anyone not already a staunch O supporter might be on the fence if asked question 5 cold. By leading in with the first 4 questions, you "remind" them of issues about which they may disagree.

The way it's formed means that the outcome will likely be skewed away from "undecided" and become more polarized.

[OneAdam12]

Ding ding ding ^^^^^^^^


Also, read question 2 then question 1 in that order. In my opinion, all polling questions should be independent and stand on their own not referencing back to any other question that would cause you to answer inconsistently.

[Unistat]

It's Newsmax. Even as right-wing as I am, I refuse to treat them as a reliable news source. It doesn't help me win any debates if my facts and arguments are based on biased sources. It's more like a series of editorials. The "poll" is just another example from that site.

I do like Chuck Norris' columns though!

[EmmaPeel]

I'm not seeing the trickery, but maybe I'm not looking close enough. The only thing I see is that some question do not have options for "unsure" or "other". I wouldn't call that trickery per se, just poor survey design.

It'ss pretty blatant trickery and manipulation when you offer no other choices but Obama.

[1042 Trooper]

Call me simple, but the questioins seem pretty straight forward to me. If you like his health plan, say so. If you like the fact that his health plan cut Medicare, say so. If you support his plan for amnesty, say so. If you want to re-elect him, say so. If you are looking at another candidate, there is a choice for that.

Where's the tricks? You guys are reading waaaaaay to much into this, IMHO

As for Newsmax, I also read the left wing Newsweek for a balance, but have read them both for years..

[Bounce]

After a little training in polling/survey creation, it's obvious what they're doing. Most training is geared toward teaching how to avoid bias. It's easy to take that training and experience and twist it to create a questionnaire that leads the reader to the conclusions you want so you can then write the article you intended to before the survey was even created.

[phillydog07]

It'ss pretty blatant trickery and manipulation when you offer no other choices but Obama.

What are you talking about? Yes.....conservative Newsmax would trick people into saying they plan on voting for Obama

[EmmaPeel]

After a little training in polling/survey creation, it's obvious what they're doing. Most training is geared toward teaching how to avoid bias. It's easy to take that training and experience and twist it to create a questionnaire that leads the reader to the conclusions you want so you can then write the article you intended to before the survey was even created.

What are you talking about? Yes.....conservative Newsmax would trick people into saying they plan on voting for Obama

I don't care who the source is or what the issue is, I don't trust INDIVIDUAL polls, regardless of whether it supports my view or doesn't. The only use I have for polling is to watch numerous polls on the same subject over time and watch their trends. Only then do I start to really pay attention to their value.

It was a poor poll, whether by ignorance or on purpose. See Bounce's post that I highlighted above.

Let me explain the "science" of polling. It's not.

Polling grew out of the field of statistics where you CAN predict a coin toss if the sampling is high enough. If you do x enough times, and if the sampling group is big enough and is valid, the science of probability is almost dead on....especially with an extremely large sample.

But that's just one side of the ART of polling. The other side is the actual questionnaire or survey. It has never been objective/scientific, and never will be.

If you wanted a purely objective poll about the presidency, you would just ask straight out, " Who are you planning to vote for in 2012?' Even that has a bias. "Who will you vote for?" is as close to pure as you get. Your next question would simply be, "Why would you vote for that person?" Not all these words that are used to bias the poll, whether intentional or not.

If you've ever read Sun Tsu's "The Art of War," there are many lessons for war today as well as for most walks of life, especially in busines and in politics. He basically counsels to use all the tools in your toolbox. Many of them have to do with psychology and deception. And if you don't think those are being used by political professionals, than I'd suggest you buy Sun Tsu....it's just a small paperback....and it will open your eyes tremendously.

[EmmaPeel]

Where's the tricks? You guys are reading waaaaaay to much into this, IMHO

.

Yea, unfortunately, Troop....I've had way too much training in the art of deception to take most anything at face value these days.

[GangGreen712]

The thing is, EVERY news source is biased. We have tricked ourselves into believing that it is possible to come at a situation from a 100% objective point of view. Even providing both sides of an argument is going to end up biased somehow. There's nothing wrong with being biased, so long as it doesn't cause you to distort or ignore the truth to a point that it is a flat out falsehood.

Like 1042 said, those statements ARE true and verifiable. Just because they are worded in a way that is barely veiling the disdain for what the Dems have done doesn't change that.

All that said, while it has a very obvious conservative bend, it's pretty straightforward and easy to interpret.

[Bounce]

And if you don't think those are being used by political professionals, than I'd suggest you buy Sun Tsu....it's just a small paperback.

it's free in eBook and for Kindle.

[SRT936]

What are you talking about? Yes.....conservative Newsmax would trick people into saying they plan on voting for Obama

No, the poll design is intended to do the exact opposite. The poll is constructed in such a way so as to lower the numbers that are pro-Obama. Its a demonstration of how easily a poll can be constructed to build a particular outcome. In this case, highly conservative (basically fringe right) Newsmax has built a poll to bolster the idea that Obama is in deep doo-doo.

Unfortunately, nearly all mainstream polls use similar tactics. Most are a bit more subtle than this one, but the bias shines through clearly.

As Emma said, polling is an off-shoot of statistics. When I was in college, I took a statistics course and the text book was entitled, "How to Lie with Numbers." Its the one field of mathematics where you start with an answer and the build the question.