This is a portion of that other discussion I referred to, which was the result of a Gallup poll vs. the MSNBC.com poll re. how President Obama is doing at his 100-day mark. The Gallup poll indicated a 56% approval rating, and the MSNBC poll asked for a "grade".
S: Online polls are notoriously inaccurate because of the people who are motivated to respond are those with strong feelings, one way or the other. Why else do you think that 38% gave Obama an A and another 38% gave him an F, leaving only 24% who responded for something in between, and where the truth probably lies.
This is particularly true with questions that are politically charged and when political extremists are guided there to vote by web sites that cater to that type.
So while I have reservations about the usefulness of polls, in general, professional polls are clearly more accurate.
AB: I might agree with that, in general, were it not for the "24% who responded for something in between, and where the truth probably lies". You see, after having already factored in the "political extremists", which in this case have ended up in deadlock (38% each), what I said was:
My math says that averages out to about a C-, rather than the B+ I've seen "reported"....
24% of 3,315,865 respondents (where the truth probably lies) is 795,808
Can you not agree that would be a more accurate reflection of the views of the American public than 1,000 or less respondents?
S: Why don't we see what MSNBC has to say about their own "unscientific poll."
Polls
Journalists and political strategists use polls to gauge what the public is thinking. The most statistically accurate picture is captured by using a randomly selected sample of individuals within the group that is being targeted, such as those likely to vote in a presidential primary election.
While a poll of 100 people will be more accurate than a poll of 10, studies have shown that accuracy begins to improve less at about 500 people and increases only a minor amount beyond 1,000 people.
So, in the case of that NBC-WSJ poll, only 2,005 adults were surveyed by the polling organizations of Peter D. Hart and Robert M. Teeter. The poll was conducted by telephone and had a margin of error of plus or minus 2.2 percentage points at the 95 percent confidence level. The confidence level means that if the same poll were conducted 100 times, each one randomly selecting the people polled, only five of the polls would be expected to yield results outside the margin of error.
In the NBC-WSJ survey, pollsters first randomly selected a number of geographic areas and then telephone numbers were generated in a way that allowed all numbers in those areas (both listed and unlisted) an equal chance to be called. Only one adult in each household was then selected to answer the poll.
Online Surveys
In contrast, msnbc.com's online surveys or Live Votes may reflect the views of more individuals, but they are not necessarily representative of the general population. And they may be even less representative of those people who are registered to vote and who do in fact vote.
To begin with, the people who respond to online surveys choose to do so they are not randomly selected and asked to participate, but instead make the choice to read a story about a certain topic and then vote on a related question.
They may be highly motivated supporters of a particular candidate who are determined to show their support for him or her in any way they can.
And while Live Votes are designed to allow only one vote per user, someone who wants to vote more than once can use another computer or another Internet account.
Live Votes are not intended to be a scientific sample of opinion. Instead, they are part of the same dialogue that takes place in our online chat sessions: a way to share your views on the news with your fellow users and with msnbc.com writers and editors.
AB: Doncha love it when a completely non-biased commentary (referring to a previous discussion) is offered as evidence? Hmmm.... Wait a minute.... I believe we already covered, for the most part, the sections you've highlighted there....
S: I don't see the bias in MSNBC admitting that their own online poll is less accurate than a poll professionally done with 1000 samples or more.
AB: Of course you don't. The statements that you did not highlight:
1.) they may be even less representative of those people who are registered to vote and who do in fact vote
2.) Instead, they are part of the same dialogue that takes place in our online chat sessions: a way to share your views on the news with your fellow users and with msnbc.com writers and editors
I particularly like the non-biased opinion expressed in #1.
Any comments here re. the "value" of professional vs. online polls?
