Pollster.com

Articles and Analysis

 

POLL: Gallup National Tracking, 1/28 through 1/30


Dems: Clinton 43, Obama 39...
Reps: McCain 37, Romney 22...

More here.

 

Comments
RS:

The Gallup poll says neither Clinton nor Obama has benefited disproportionately from Edwards' withdrawal.
I don't know - Edwards has dropped 4 points from the last poll. Clinton is up 1, Obama is up 3.

Making trends of ONE data point... :-)

____________________

Martin:

Let's wait and see what tomorrows Gallup poll suggests. Then we could actually draw a line from dot to dot.... ;-)

____________________

Paul:

If you look at the state polls since South Carolina, the gap between Clinton and Obama has narrowed or in some cases Obama has gone into the lead. National polling also shows a narrowing. Question is extent to which there is enough time before Tuesday February 5 for Obama to catch Clinton. The important point which most in the media are not focusing on is the questions of delegates. None of the Democratic Super Tuesday states are winner takes all. So if Obama takes say 48% and Clinton takes 52%, the difference of 4% will be relatively small in terms of delegates. Based on delegates and a close February 5 contest, and considering prior state delegate awards, I do not see how the Democratic nomination is wrapped up Tuesday.

____________________

RS:

@Paul:

This is of course assuming one candidate does not make a HUGE mistake pre-Feb 5. Example, Obama doesn't shake Clinton's hand... ;-)

____________________

vh:

Hey you Obama idiots who are tracking pt. by pt. All these polls have error of more than +-3%.
So all your tracking is waste of time which all of you see to have lot of it.
LOL

____________________

RS:

@vh:
Thanks for making the better point - the race is (almost) tied!! And guess who has the big mo'?

:-)

____________________

dhcobra:

After yesterday's debate I wouldn't be suprised if Clinton & Obama Team up as the Dem ticket.

A Clinton/Obama ticket would be unbeatable even if a third party spoiler joined the race.

____________________

FlyOnTheWall:

Hey, vh. The margin of error is a confidence interval. In layman's terms, we can be highly confident that the actual level of support lies within, in this case, 3 points of the stated level. (At least, in theory. See the series of splendid posts on polling errors this cycle for a darker view.)

But it's worth noting that not all results within that range are equally probable. That is, the most probable actual level of support is the one listed in the poll. The further away you deviate from that number, the less probable it becomes, until the result becomes so improbable that it lies outside the interval.

That's a long-winded way of saying that even movements or leads smaller than the margin of error can matter, and are worth tracking. If two candidates poll 49% and 47% in a +/-3% poll, we can say that they are statistically tied - that is, that the poll in question has not established with sufficiently high confidence that one is outperforming the other. But they are not actually tied - it remains somewhat more probable that the candidate with 49% is actually ahead. There is, after all, some actual level of support in the general population that the poll approximates, and we can try to establish the odds that the poll has done so correctly.

Does that help?

____________________

Jeff:

To put a fine point on it, in a poll showing a 2% lead with a 3-point margin of error, there is a 75% probability that candidate A is genuinely ahead of candidate B (and not ahead simply because of sampling error), according to some research I saw from some professors of mathematics and statistics at Cal State Chico.

Dim political reporters almost always refer to a lead within the margin error as "a statistical tie," which shows almost no grasp of statistics (or ties, for that matter). You are under no obligation to repeat their mistake, vh, and as most of the visitors here are pretty seriously interested in survey research, there are much better sites to toss around abusive political remarks.

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR