[Winstonm]

Quote

Poll: News media’s credibility plunges to all-time low
By Agence France-Presse

Quote

WASHINGTON — Public trust in the US media is eroding and increasing numbers of Americans believe news coverage is inaccurate and biased, according to a study released on Monday.

Quote

Just 29 percent of the adults said news organizations generally get the facts straight.

Sixty-three percent said news stories are often inaccurate.

Sixty percent said the press is biased.

Seventy-four percent said news organizations tend to favor one side in dealing with political and social issues.

Seeing that this was a news release, I'm giving it a 50/50 shot at being accurate.

[MattieShoe]

Haha, it's a strange world we live in, where Jon Stewart is the most trusted newsman in the country.

Link to poll

Incidentally, I trust Jon Stewart too. I don't know why exactly -- Maybe it's just that their motivations are crystal clear. I already know he's a liberal personally, but the show's goal is just to make people laugh and sell advertising time, at either party's expense. :-)

[blackshoe]

Trust, but verify.

[matmat]

MattieShoe, on Sep 20 2009, 07:02 PM, said:

Haha, it's a strange world we live in, where Jon Stewart is the most trusted newsman in the country. 

Link to poll

Incidentally, I trust Jon Stewart too.  I don't know why exactly -- Maybe it's just that their motivations are crystal clear.  I already know he's a liberal personally, but the show's goal is just to make people laugh and sell advertising time, at either party's expense.  :-)

thats some quality poll, with, say, Montana registering all of 7 votes.

Clearly a scientific poll with an unbiased base sample ,etc.

here's another:
http://www.timepolls.com/hppolls/archive/p...esults_458.html

[barmar]

To be fair, it says right there on the page "Note: Poll results are not scientific".

[jdonn]

barmar, on Sep 22 2009, 04:08 PM, said:

To be fair, it says right there on the page "Note: Poll results are not scientific".

I saw that same qualifier on a cnn.com poll last week that asked "which is your favorite dinosaur" with the options "Tyranosaurus Rex, Brontosaurus, Triceratops, Barney". Which is good because otherwise I would have assumed it was a scientific poll.

[Winstonm]

jdonn, on Sep 22 2009, 04:19 PM, said:

barmar, on Sep 22 2009, 04:08 PM, said:

To be fair, it says right there on the page "Note: Poll results are not scientific".

I saw that same qualifier on a cnn.com poll last week that asked "which is your favorite dinosaur" with the options "Tyranosaurus Rex, Brontosaurus, Triceratops, Barney". Which is good because otherwise I would have assumed it was a scientific poll.

Yeah, but everyone knows how gullible you are.

[MattieShoe]

I hate it when the word scientific is used like this. what makes a poll "scientific"? Is it biased? Certainly, all poll sare biased. No special effort was made to unbias this one, it's just a silly internet poll. If internet polls decided anything, Ron Paul would be president.

Still, that doesn't mean the results aren't interesting.

[jdonn]

MattieShoe, on Sep 23 2009, 09:23 AM, said:

I hate it when the word scientific is used like this. what makes a poll "scientific"? Is it biased? Certainly, all poll sare biased. No special effort was made to unbias this one, it's just a silly internet poll. If internet polls decided anything, Ron Paul would be president.

Still, that doesn't mean the results aren't interesting.

You seem to be confusing scientific with perfect.

[helene_t]

MattieShoe, on Sep 23 2009, 03:23 PM, said:

I hate it when the word scientific is used like this.  what makes a poll "scientific"?

A "scientific" poll may be a sort of buzz word sometimes but it is not completely meaningless.

To make a poll scientific, you need to control its bias. You may not be able to eliminate all kinds of bias, but at least the poll should be based on an established, well-described methodology that allows critical readers to assess the magnitude of the bias.

For example, if you have polled random people found in the white pages, you bias against potential respondents likely not to be listed, to refuse to answer, or not to be reachable. Say you specify that you tried each respondent up to 3 times, two weekday evenings between 8 and 9 and one Sunday between 11 and 17, or at a time recommended by another member of the household. Also, you specify in the protocol exactly how the question was phrased and how possible answers were categorized. You must report how many people were selected but did not participate for various reasons. Then readers can judge the quality of your poll, for example by looking at studies that compare the results from polls using the described method to polls using other methods.

In contrast, just setting up an internet site where anyone can vote is subject to completely uncontrolled bias.

[barmar]

And it's not like we don't allow biased results to strongly influence our lives. Voting in general elections has no controls against bias, but we still use this method to fill in much of our government. There's very likely to be a strong correlation between political attitudes and interest in voting (you're probably more likely to vote if the opposition is in power, because you want to throw the bums out, while their constituents may be more complacent).

[kenberg]

Poll numbers can be interpreted in various ways. browsing through the Time polls from matmat's link I found the question

Should Michael Jackson's grave site be open to the public?

In North Dakota, 75% said yes. But the 75% consisted of 3 of the 4 responses. I prefer to interpret this result as saying that the remaining 641,477 residents don't give a hoot. This speaks well of them.

[barmar]

kenberg, on Sep 24 2009, 01:44 PM, said:

Poll numbers can be interpreted in various ways. browsing through the Time polls from matmat's link I found the question

Should Michael Jackson's grave site be open to the public?

In North Dakota, 75% said yes. But the 75% consisted of 3 of the 4 responses. I prefer to interpret this result as saying that the remaining 641,477 residents don't give a hoot. This speaks well of them.

I suppose you could look at it that way, but a more realistic interpretation would be that most of the rest weren't even aware that the question was asked.