Mark Blumenthal | March 4, 2008
Topics: 2008 , Barack Obama , CBS , CNN , Exit Polls , Hillary Clinton , Mark Lindeman , MSNBC
Ok, another Tuesday night and another election night. Polls will close at 7:00 p.m. (Eastern). at 7:30 p.m. in Vermont and 9:00 p.m. in Rhode Island and Texas. Exit polls tabulations will be posted by the networks at these links:
Carrying on with our "live blogging" tradition, I'll post what seems relevant here on what we can learn from the non-leaked exit poll information tonight. Updates will follow in reverse chronological order -- all times Eastern.
11:32 - Using a far less sophisticated extrapolation, the current Ohio estimate looks to be 55% Clinton, 43% Obama. Quite a shift from the initial four point estimate.
11:25 - Mark Lindeman is signing off to get some sleep. I'll post what seems relevant, but for now, our estimate extrapolator is offline. Thanks for your help Mark!
11:11 - The tabulations for Vermont and Rhode Island have updated. Rhode Island's estimate now shows a 59% to 40% Clinton lead (was 51% to 48% before the update -- quite a shift). Vermont shows a 60% to 38% Obama lead.
10:05 - I'm going to be offline for about 30 minutes while I relocate. Until then, look for Lindeman's updates in the comments.
9:40 - Promoted from the comments, from my very able helper Mark Lindeman:
In case anyone is wondering, that wasn't a typo about RI. The exit poll tab, consistent with the early leak, shows a very close race -- but the networks have seen enough votes to make a call. It's not news that the exit polls have often (but not consistently) overstated Obama's performance.
9:37 - From the Texas Democratic crosstabs, a preliminary look at the other set of numbers I've been obsessing over for the last few days (numbers from 2004 and 2000, in that order, in parentheses):
- 57% Female (52%)
- 30% Latino (24%)
- 19% African American (21%)
- 16% Age 18-29 (9%)
- 44% Age 18-44 (34%)
- 55% Age: 18-50 (47%)
- 13% Age: 65+ (19%)
- 33% Independent/Republican (26%)
- 43% College degree (37%)
- 61% Income: $50K+ (49%)
One extra caution on Texas: From what I understand, the estimates are especially subject to change because they weighting on early vote vs election day vote, and the geographic mix (which helps determine the racial composition) really requires hard count to get right. And if that last sentence goes over your head, the point is, the numbers above are not set in stone.
9:21 - [Sorry this got garbled a few moments ago -- finishing the thought now]: Caro asks "Are the early votes (pre-election) included in the exits?" In Texas the exit pollsters would have conducted a telephone survey of early voters that they fold into the interviews conducted at polling places. They have done telephone interviews with early voters in Texas before. With the number of early votes cast in Ohio, I assume that an absentee vote telephone survey was conducted there as well, but I'm not certain.
9:10 - More estimates crunched by Mark Lindeman: In Rhode Island, Clinton 51%, Obama 48. Among Texas Republicans, McCain 49%, Huckabee 38%.
9:02 - The current cross-tab estimate for Texas shows 50% Clinton, 49% Obama. See the usual caveats below
7:45- From the Ohio Democratic crosstabs, a preliminary look at the numbers I've been obsessing over for the last few days (numbers from 2004 and 2000, in that order, in parentheses):
- 59% Female (52%, 60%)
- 19% African American (14%, 17%)
- 15% Age 18-29 (9%, 8%)
- 44% Age 18-44 (32%, 36%)
- 54% Age: 18-50
- 31% Independent/Republican (29%, 24%)
- 37% College degree (37%, 27%)
- 56% Income: $50K+ (49%, 45%)
Keep in mind that these numbers, like all in the current tabulations, are preliminary and will likely change over the course of the night.
7:35 - As polls close in Ohio, the posted tabulations show 52% Clinton, 48% Obama. Again (can't say this enough) these are preliminary estimates that will grow more accurate as the night wears on. See the 6:59 entry for caveats.
7:22 - Pollster reader Thatcher has reposted some leaked mid-day exit poll results found on Huffington Post. For those who want to speculate about the meaning of those numbers -- and that's everyone, right? -- I'd recommend my exit poll tips last posted on February 5 and especially the "comparable" leaked numbers posted at about this time that day. Bottom line: a percentage point or two in either direction on a mid-day exit poll doesn't mean much.
7:10 - A few minutes ago, MSNBC's Nora O'Donnell reported two key statistics from the very preliminary exit poll tabulations from Texas and Ohio. In Ohio, 22% of Democratic primary voters described themselves as independent and 10% as Republican, that compares to 24% and 2% respectively in 2004. In Texas, 24% in the early tabulations are independent and 10% Republican (which compares to 20% and 5% respectively in 2004). For what it's worth, the combined non-Democrat percentage in Ohio is 8 to 15 percentage points higher than any of the pre-election polls that released party ID results. In Texas, that number is on the high side of what public polls were showing.
7:00 - MSNBC projects Barack Obama the winner, but Keith Olbermann tells us that "we do not have a number" yet for Vermont. That's true -- they do not have a number they consider "air-worthy." However, they have posted preliminary cross-tabulations on the MSNBC web site, and those currently indicate an estimate of 64% for Obama, 34% for Clinton. See the caveat below -- these estimates are preliminary and will become more accurate as the evening progresses.
6:59 - Shortly after the polls close in each state, our friend Mark Lindeman will report the extrapolated overall vote estimate used to weight the exit poll cross-tabulations. These estimates begin as a mashup of pre-election polls and the interviews exit polls conducted at polling places and over the phone (with early voters) by the networks. These estimates improve, becoming more accurate over the course of the night. Click here for the usual caveats on how these numbers are derived and how they improve over the course of the evening.