WV Secretary of State
Posted on 05/08/2012 7:38:56 AM PDT by chippewaman
On Tuesday, May 3rd, West Virginia voters will go to the polls to nominate candidates for federal and state offices. The Democratic presidential primary that day will be a head-to-head contest between President Barack Obama, and Keith Judd, a felon currently serving time in a federal prison for making threats on a college campus. And according to a model developed through multiple regression analysis, I believe there is a very good chance that the resident of the White House, President Obama, will lose several West Virginia counties to a resident of a prison cell in Texas. Background
West Virginia was one of Barack Obama's weakest states during the 2008 primary season; he lost there by a thirty-nine point margin to Hilary Clinton (67%-28%). And while West Virginia is not a state that has been polled extensively this year, the most recent Public Policy Polling from the fall of 2011 recorded Obama's approval rating among West Virginia Democrats as being only 45%.
All of these facts raised the following question in my mind. Could West Virginia's Democratic primary be competitive? Or, at a minimum, could Keith Judd earn a significant number of votes? To try to figure that out, I looked at the primary results in two states where President Obama has faced primary challengers in 2012 and where he suffers from pockets of unpopularity among registered Democrats (Oklahoma and Louisiana).
Breaking down the results on a county-by-county basis and testing certain variables, I discovered three factors that heavily influenced the percentage of the Democratic primary vote earned by President Obama in any given county this year against other candidates:
(1) The percentage of the vote earned by Obama in the 2008 primary. (Obviously, Obama doing well in the 2008 primary has a positive impact on his 2012 performance).
(2) The percentage of the vote earned by Obama in the 2008 general election. (Again, Obama doing well in the 2008 general election has a positive impact on his 2012 performance).
(3) The percentage of the electorate registered as Democrats. (The percentage of the electorate registered as Democrats actually has a negative impact on Obama's 2012 performance. This is because counties with a high percentage of registered Democrats tend to have more conservative Democrats).
While I experimented with other variables (e.g., non-Hispanic white population, percentage of adults with college degrees), those variables were not statistically significant and did not enhance the explanatory power of the model given the presence of the other variables.
In any event, using the three variables listed above, multiple regression analysis produced a formula to explain the variance in Obama's 2012 vote among various counties in Oklahoma and Louisiana. This formula correctly predicted in 75 out of Oklahoma's 77 counties whether Obama would break 50% in a county. This formula also correctly predicted in 59 out of Louisiana's 64 parishes whether Obama would break 50% in a parish. Moreover, in the multiple regression analysis, each of the three variables was quite statistically significant, and the formula explained the substantial majority of the variation of Obama's vote among counties.
Using this formula, I then turned to see what it forecasts will happen in the West Virginia primary. Admittedly, there is one notable difference between the West Virginia primary and those in Oklahoma and Louisiana. In Oklahoma and Louisiana, President Obama faced multiple challengers. In West Virginia, however, he is only facing one challenger. I don't think, however, that this fact, in and of itself, should make much difference for purposes of my analysis. In each contest, Democratic voters are almost certainly deciding whether to vote for or against President Obama, and the fact that there are also three non-Obama names rather than one one-Obama name shouldn't have much impact on the percentage of the vote that President Obama receives.
County Projections
Simply put, Keith Judd has a very good chance of defeating President Obama in some West Virginia Counties from his prison cell in Texarkana, Texas. Specifically, my formula projects that Barack Obama will lose thirteen West Virginia counties to Keith Judd on May 3rd: Clay; Gilmer; Hardy; Lewis; Lincoln; Logan; Mercer; Mingo; Ritchie; Summers; Wayne; Wirt; and Wyoming.
Most of these counties (7) are in Southern West Virginia's coal country while five are in North-Central West Virginia and one (Hardy) is on the western outskirts of the Eastern Panhandle. According to the formula, Judd's margin of victory will be in double-digits in five of these counties (Lincoln, Logan, Mingo, Wayne, and Wyoming). Indeed, in Wayne County, Obama is projected to lose by a two-to-one margin. Judd's margin of victory is projected to be single digits in the other eight counties. In particular, Judd's margin of victory is projected to be fewer than five points in Mercer, Ritchie, Summers, and Wirt so these counties could easily go the other way.
The good news for Barack Obama is that my model forecasts that he will carry West Virginia's other forty-two counties. However, several of them could go the other way. Specifically, his margin of victory is projected to be fewer than five points in the following eight counties: Boone; Calhoun; Doddridge; Monroe; Pendelton; Raleigh; Tucker; and Webster. As a result, I wouldn't be surprised to see him lose any of these counties.
The model projects that Obama will perform best in the easternmost part of the Eastern Panhandle, carrying Berkeley County with 72% of the vote, and Jefferson County with 82%. It also forecasts that Obama will win Kanawha County, West Virginia's most populous county, comfortably with 68% of the vote.
Statewide Projection
Looking to the total statewide vote, my model forecasts that Barack Obama will defeat Keith Judd by a margin of 59%-41%. His strength in the state's two panhandles and more populous counties offsets his projected weakness in coal country and some other rural counties, allowing him to win the state. (By means of comparison, President Obama received 57% of the vote in the Oklahoma Democratic Primary, and 76% of the vote in the Louisiana Democratic Primary.) My model, however, does a better job of producing correct results at the county level than at the statewide level. In Oklahoma and Louisiana, it understated Obama's statewide vote by 4-5 points, due to differences in voter turnout throughout the state. As a result, I would not be at all surprised if Obama's vote total ended up at around 63% or 64% to Judd's 36% or 37%.
Caveat
Were President Obama's opponent in the West Virginia primary an average person with no name recognition, I would be fairly confident in the forecast produced by my model. In fact, I think that it would be extremely unlikely for Keith Judd not to capture at least one-third of the vote.
If, however, it becomes widely known among Democratic voters that his opponent Keith Judd is currently a federal prisoner, it is quite possible that Obama's percentage of the vote will be boosted substantially. While I doubt that many people who were previously likely to cast protest votes for Obama's opponent would switch to Obama, they may simply not cast a ballot for President and just vote in other contests, thus increasing Obama's percentage. (On the other hand, it is also possible that those voters who are inclined to vote against Obama will not care about the identity of his opponent since their motivation is to protest Obama. Indeed, one might say that it is a more powerful protest to vote for a convicted felon over Obama as opposed to an average no-name individual.)
Conclusion
It is almost certain that President Obama will win West Virginia's Democratic Primary on May 3rd. On the other hand, I do believe that it is more likely than not that he will lose some counties to federal prisoner Keith Judd and that Judd will receive a substantial protest vote statewide that could prove embarrassing to the President.
Could be an interesting night in WV. If prisoner ends up winning some WV counties I doubt it will be reported on CNN or MSNBC. How embarassing!
I recall that Lyndon LaRouche used to rack up some significant vote totals in some Democrat primaries but the Dems and their media allies would cast a pall of silence around them out of embarrassment.
So also will it be with Keith Judd, who could win an even more embarrassing number of votes.
May 3rd? Where are the results?
Obama is so disliked in the Third Congressional District (southern coalfields) that he may be upset by the prisoner in those counties. I recall Obama losing to Hillary Clinton 88-8 in one of those counties in the 2008 Dem primary.
Article is wrong. WV primary is today May 8.
When you read the Charleston,WV papers, you see alot of disgust for Gov. Tomblin and Sen. Manchin not supporting Obama. I don’t see how anyone in WV could vote for Obama, but they are so ingrained in the unions and the Democratic party that common sense goes out the window.
Judd got my vote.
The Charleston media has always been liberal. Tomblin and Manchin have distanced themselves from Obama in order to get re-elected. WV is a strange state politically. It’s what I call a DINO state. There are a lot of Old South conservatives, who call themselves Democrats. They haven’t realized that it’s not 1955 anymore. The national Democratic platform has left them.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.