Monthly Archives: November 2014

2014 midterm election: results vs. predictions

As of November 14, there are still a handful of Congressional races yet to be called. Nonetheless, if those that are currently leaning toward the GOP end up in the Republican column, we’ll begin the 114th Congress with 247 Republicans and 188 Democrats in the House, a pickup of 13 seats for the GOP. Assuming that Mary Landreiu loses reelection in the runoff election on December 6, the Senate will have 54 Republicans and 46 Democrats, a pickup of 9 seats in the Senate for the GOP.

In October, PS: Political Science and Politics published a collection of forecasts of the 2014 midterm elections. These were forecasts done by political scientists who make predictions based on election “fundamentals” such as presidential approval and economic conditions, and done several months before the election took place. (Notably, these models do not include information on things like campaign spending, candidate competence/appeal, etc.) The average (median) prediction was that the Republicans would pick up 14 seats in the House and 5 or 6 seats in the Senate.

All in all, not a bad showing for the science of political science election forecasting in 2014.

Predicted result Actual result Margin of error (difference / total seats)
House + 14 GOP + 13 GOP 0.2%
Senate + 5 or 6 GOP + 9 GOP 3.5%

2014 Boyle County Exit Poll: How accurate are the results?

Here is a comparison of the 2014 BCEP results for various races and the actual results from the Kentucky Secretary of State‘s website:

Race Actual results Survey results Difference Survey results with sample weighting Difference
McConnell 56.5% 50.2% 6.3% 51.8% 4.7%
Grimes 40.5% 47.3% 6.8% 46.1% 5.6%
Patterson 3.0% 2.4% 0.6% 2.1% 0.9%
Guthrie 62.5% 58.4% 4.1% 59.2% 3.3%
Leach 37.5% 41.4% 3.9% 40.6% 3.1%
McKinney 55.0% 59.9% 4.9% 61.4% 6.4%
Harmon 45.0% 40.1% 4.9% 38.6% 6.4%
Mike Perros 51.0% 43.8% 7.2% 45.4% 5.6%
Paige Stevens 49.0% 56.2% 7.2% 54.6% 5.6%
Steve Becker 20.3% 18.8% 1.5% 19.6% 0.7%
Lowery Anderson 15.9% 17.7% 1.8% 18.4% 2.5%
Susan Matherly 25.7% 24.4% 1.3% 23.8% 1.9%
Paige Matthews 23.6% 24.7% 1.1% 23.9% 0.3%
Elaine Wilson-Reddy 14.6% 14.4% 0.2% 14.3% 0.3%
Rick Serres 17.0% 17.7% 0.7% 17.7% 0.7%
Denise Terry 17.1% 15.9% 1.2% 15.5% 1.6%
Kevin Caudill 20.1% 20.4% 0.3% 20.7% 0.6%
Kent Mann 13.2% 12.3% 1.0% 12.3% 1.0%
J. H. Atkins 19.8% 20.4% 0.6% 19.7% 0.1%
Buck Graham 12.3% 13.4% 1.1% 14.2% 1.9%

A reminder that a sample weighting was applied to figures reported on this blog, which is a standard procedure used to correct for differences in how different demographic groups respond to the initial invitation to take the survey.

In general, it seems that the exit poll sample tended to overstate support for Democratic candidates in the partisan races as well as the implicitly liberal-leaning candidate in the non-partisan Danville mayoral election by anywhere from 3%-6%.[1] When it comes to the Danville city commission and school board, however, the exit poll sample was only about 1% off from the actual final results. The average difference between the exit poll sample and the actual sample for all races is 2.8% which is reduced slightly to 2.7% once a sample weighting procedure is applied.

Given that the results vary by about an average of 2.5%-3%, we can confidently assume that the other responses from the survey questions are likely somewhere in the same ballpark. This range is well within the standard accepted margin of error of 3% for most national surveys and professional polling firms.

[FN1] There are a number of explanations that can account for this, including the possibility that people were slightly less likely to accurately report their voting patterns for these higher-profile offices on the survey form. However, it is ultimately impossible to definitely verify this one way or the other.

Boyle County Exit Poll 2014: Minority religious affiliation of Boyle County voters

Among many other socio-political and demographic questions, the 2014 Boyle County Exit Poll asked respondents to report their religious affiliation. As reported previously, Evangelical Protestants made up 38% of the sample, with 27.5% Mainline Protestants, 0.7% Black Protestant, 9.9% Catholic, 8.6% religiously unaffiliated, 4.2% “atheist/agnostic,” 5.7% “other” religion.

Here is the breakdown of what respondents wrote in the “other” religious category. The first column indicates what was written, the second reports exactly how many individuals wrote that answer (or some close variant of it).

“Just believe in God and love” 1
Adventist 1
Baptist 4
Buddhist 1
Calvinist 1
Catholic 2
CBF Baptist (not evangelical) 1
Christian 32
Christian Buddhist 1
Church of Christ 2
Church of God 1
Deist 1
Greek Orthodox 1
Independent Baptist 1
Jewish 6
Lutheran 2
Messianic Jew 2
Monotheistic Pagan 1
Mormon/LDS 3
Native American 1
Nazarene 1
None of your business 1
Pagan 1
Pentecostal 2
Presbyterian USA 1
Spiritual 3
Unitarian Univeralist 1

It was reported previously that about 7% of the adult population of Boyle County, Kentucky was included in this survey sample. Thus, if you multiply the above figure by about 14, you’ll get a very approximate “ballpark” estimate of about how many people in Boyle County, Kentucky self-identify with that particular religious affiliation.

Midterm Election Review on WEKU’s “Eastern Standard”

More discussion on the midterm election results: my interview on WEKU’s “Eastern Standard” program hosted by John Hingsbergen. This program also features Dr. Joseph Gershtenson from EKU and Nick Storm from CN2.

“Flameside Chat”: Debriefing the 2014 midterm election

On the evening of Wednesday, November 5th, Dr. Dan Stroup, Dr. Lia Rohr, and myself sat down to participate in an hour-long “Flameside Chat” debriefing the results from Tuesday’s midterm election. Enjoy!

Boyle County Exit Poll 2014: Survey response rate

We had a total of 1,684 individuals who agreed to take our exit poll survey. There were also 1,868 individuals who were asked to take the survey but who declined to do so. Thus, we asked a total of 3,552 people to take the survey and 1,684 of them agreed, leaving us with a response rate of 47.4%.

This is down 3% from the response rate in the 2012 exit poll of 50.4%.

We can also note that 9,720 people voted on Election Day in Boyle County. Thus, our 2014 BCEP exit poll includes 17.3% of all voters on Election Day, and about 7% of all adults over age 18 in Boyle County (voters and non-voters).

Boyle County Exit Poll 2014: Voting patterns in the Danville ISD School Board Election

Below are some of the results for the Danville ISD school board election from the 2014 Boyle County Exit Poll.

As per the KY Secretary of State’s website, Susan Matherly received 25.7%, Paige Matthews received 23.6%, Steve Becker received 20.3%, Lowery Anderson received 15.9%, and Elaine Wilson-Reddy received 14.6% of the final vote.

The following table presents the percentage of voters who selected one candidate for who also voted for another candidate. To interpret: find the name going across and then another name going down. The number indicates the percentage of people ACROSS who also voted for the person going DOWN. So for example, 42.8% of Anderson voters also voted for Becker.

Becker Anderson Matherly Matthews Wilson-Reddy
Becker 100.0% 42.8% 41.8% 40.9% 43.1%
Anderson 38.9% 100.0% 46.7% 42.0% 36.2%
Matherly 49.2% 60.6% 100.0% 61.6% 57.6%
Matthews 49.5% 55.9% 63.4% 100.0% 52.2%
Wilson-Reddy 30.6% 28.2% 34.7% 30.5% 100.0%

We see that some of the strongest relationships are between Matthews/Matherly and Anderson/Matherly.

The survey also asked “do you think students in [the Danville ISD] school district perform: better/about the same/below state average?” In general, those who were tended to think that students were performing about the same or below state average tended to be slightly more likely to choose Becker, Matthews, and Wilson-Reddy.

They survey also asked “What are the chief factors that influenced your vote for Danville Board of Education?”

  • Those who answered “candidate qualifications” were slightly more likely to select Becker and Anderson.
  • Those who answered “candidate stands on issues” were more likely to vote for Becker.
  • Those who answered “know candidate personally” were more likely to vote for Anderson, Matherly, and Matthews.
  • Those who answered “recommendations from others” were slightly more likely to select Anderson, Matherly, Matthews, and Wilson-Reddy.
  • Those who answered “number of yard signs” were more likely to vote for Matthews.
  • Those who answered “location/placement” of yard signs were more likely to vote for Anderson, Matthews, and Wilson-Reddy, and less likely to vote for Becker.

The survey further asked “How well informed did you feel in your vote for Danville Board of Education?” Those who felt more informed were more likely to select Anderson, Matherly, and Matthews.

Voters who preferred Mike Perros for Danville mayor also were more inclined to prefer Becker but less likely to prefer Matherly or Matthews.

There were also some interesting demographic patterns:

  • Women were more likely to select Matherly and Matthews and men were more likely to select Becker and Anderson.
  • Self-described liberals tended to prefer Matthews and Wilson-Reddy while self-described conservatives tended to prefer Becker.
  • Older voters tended to prefer Becker and Anderson while younger voters preferred Matthews and Wilson-Reddy.
  • Those with higher incomes tended to prefer Wilson-Reddy while those with lower incomes tended to prefer Becker.
  • Non-white voters were slightly less likely to vote for Wilson-Reddy.

It should be noted again that these are all weak to moderate relationships. None of these patterns are “iron clad” or 100% predictive to any degree. Nonetheless, these were some of clear patterns that emerged in choices for Danville School Board in 2014.

Finally, respondents were also asked if they were a “parent/guardian of a student in the Danville school system.” Interestingly, there was no discernible relationship between this question and voting patterns for any of the candidates for school board.