Why did the Networks Cancel Exit Polls in 19 States?
Oct. 4, 2012
The decision to eliminate 2012 election exit polls in 19 states by the National Election Pool is a blow to Election Integrity. Unadjusted state exit poll data have been a major component in calculating exit poll discrepancies.
Of course, we don’t get to see the unadjusted exit poll numbers until months or years after the election. But having the data for just 31 states means that it will no longer be possible to compare the total weighted average of the state polls to the official recorded share.
The full set was required in the 1988-2008 unadjusted state-exit-polls statistical reference to show that in the six presidential elections, the Democrats won the average unadjusted state and national exit polls by a 52-42% margin. Their recorded margin was just 48-46%.
These states will be excluded: Alaska, Arkansas, Delaware, District of Columbia, Georgia, Hawaii, Idaho, Kentucky, Louisiana, Nebraska, North Dakota, Oklahoma, Rhode Island, South Carolina, South Dakota, Tennessee, Texas, Utah, West Virginia and Wyoming. This worksheet/graph displays 2008 recorded vote and exit poll statistics for each state.
Although not battleground states, they are still required to calculate the total weighted average National Vote Share. In fact, the unadjusted exit polls in these states showed that Obama did much better in the polls than the recorded vote..
McCain led Obama by 55.3-43.4% (3.4 million) in the 19 states, but only by 51.2-47.3% (1.1 million) in the corresponding exit polls. Obama actually won the exit polls in Alaska, Nebraska and Georgia. He was exactly tied with McCain in South Dakota.
Romney may assume that Obama will win the overall popular vote, just like Bush assumed Kerry would in 2004. Bush could not have a repeat of the 2000 election in which Gore won the popular vote. Bush needed votes in close battleground states like Ohio in order to capture the all-important electoral vote, but he also needed to win a popular vote “mandate”. Given the exit polls in vote-rich Democratic states like New York and California, Kerry’s margins were reduced by one-third. No one even considered that the official counts were fraudulent since Kerry won the states in landslides. The focus was on Ohio.
Consider Texas with its 38 electoral votes and 8 million popular votes. Obama lost the Texas exit poll by 6.1% (52.3-46.2) and the recorded vote by 11.8% (55.4-43.6), a 460,000 difference in margin. With a large and growing Hispanic vote, could Obama have a chance of actually winning Texas?
Without having the exit poll, it will be more difficult to estimate if the official, recorded vote is legitimate. If Romney’s true Texas margin is padded by 500,000, why would anyone be suspicious without an exit poll?
But we do have the True Vote Model , which can calculate the net defection of returning Obama voters that would be required to match the recorded vote. Maybe the TVM will provide some clues in lieu of the exit polls.
In 2008, in order to match the Texas recorded vote, McCain needed 21% of returning Kerry voters while Obama had 15% of returning Bush voters – a 6% net defection of Kerry voters.
Is the corporate media preparing for 2004 redux? The pollsters will continue to provide the National Exit Poll, a subset of the state polls which includes just 20% of the state respondents. But as it is standard operating procedure, the poll will be forced to match the recorded vote. It’s a moot point, since we are not going to see the unadjusted, pristine poll numbers until long after the election, if then.
The Director of Elections for ABC News, a member of the consortium that runs the exit poll, said the aim “is to still deliver a quality product in the most important states,” in the face of mounting survey costs, partially due to the continued rise in the number of cell phones which increases the cost of phone surveys.
He says that “the decision by the National Election Pool — a joint venture of the major television networks and The Associated Press — is sure to cause some pain to election watchers across the country”. He’s right about that.
But how much is transparency in our elections worth?
Election Model Forecast; Post-election True Vote Model
2004 Election Model (2-party shares)
Kerry 51.8%, 337 EV (snapshot)
State exit poll aggregate: 51.7%, 337 EV
Recorded Vote: 48.3%, 255 EV
True Vote Model: 53.6%, 364 EV
2008 Election Model
Obama 53.1%, 365.3 EV (simulation mean);
Recorded: 52.9%, 365 EV
State exit poll aggregate: 58.0%, 420 EV
True Vote Model: 58.0%, 420 EV
2012 Election Model (2-party shares)
Obama 51.6%, 332 EV (Snapshot)
Recorded : 51.6%, 332 EV
True Vote 55.2%, 380 EV