Selevtive reporting of poll numbers

Rush quoted the poll numbers form a website called Commentary Magazine. I'll share the numbers also.

What's interesting is that Rush shared the numbers - but did not give all the news. Right after those numbers the article goes on to say, "Not all the news was bleak for the president." I wonder why he didn't see fit to mention that?

As Alana wrote earlier today, according to the latest McClatchy-Marist poll:

Only 37 percent of registered voters approve of President Obama’s handling of the economy, his lowest rating ever.
By nearly 2-1 (61 percent v. 32 percent) voters disapprove of how he’s handling the federal budget deficit.
Fifty-eight percent of voters disapprove of Obama’s handling of the economy, comprising 60 percent of independents, 31 percent of Democrats and 91 percent of Republicans.
Not all the news was bleak for the president.

Fifty percent of voters said they had favorable impressions of him versus 44 percent who didn’t. By 2-1, Americans said today’s economic conditions mostly were something the president inherited rather than the result of his own policies. And overall, 45 percent said they approved of the job the president is doing, while 47 percent disapproved.

Overall, though, these numbers continue an alarming trend for the president.

“It’s a real caution sign … the four-year lease on the White House is very much dependent on how people end up looking at the economy,” said Lee Miringoff, director of the Marist Institute for Public Opinion, which performed the survey.

If Miringoff is right, then Obama’s lease on the White House may very well end on the first Tuesday in November 2012.

But of course what these articles reporting poll numbers never say (they just give a link to another page - how many people follow that link, I wonder) is that they are making these predictions based on talking to:

The poll of 1,003 adults, including 801 registered voters, was conducted June 15-23. The margin of error is plus or minus 3.0 percentage points for the entire sample and plus or minus 3.5 percentage points for registered voters.

I'm sorry, but polling a measly 1,000 people is not enough to make these kinds of assumptions! Not when there's bloodly 100 gazillion people in the US! Did these pollsters talk to blacks? Hispanics? People in poor towns who get their income from welfare?

It's time pollsters started talking to at least 10,000 people before even attempting to analyze what poll results mean, and it's time that articles reporting on these poll results actually put the number of people talked to in the first paragraph of the article. Space might be limited in paper-newspapers, but online newspapers have plenty of room to include this information.