Category Archives: Exit Poll

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%
AVERAGE DIFFERENCE 2.8%   2.7%

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.

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.

2014 Boyle County Exit Poll: patterns in Danville City Commission voting

Here’s another way of looking at the results of the 2014 Danville City Commission election. The following table presents the percentage of voters who selected one candidate for mayor/city commission who also voted for another commission 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, 64.1% of Terry voters also voted for Caudill. Another example: 51.9% of those who voted for Perros for mayor also voted for Terry for commissioner.

Perros Stevens Serres Terry Caudill Mann Atkins Graham
Serres 60.1% 53.6% 100.0% 59.2% 57.1% 57.1% 53.1% 56.9%
Terry 51.9% 48.4% 52.2% 100.0% 49.2% 50.5% 47.9% 46.0%
Caudill 59.2% 68.7% 65.7% 64.1% 100.0% 64.2% 71.8% 61.9%
Mann 46.3% 34.4% 39.5% 39.6% 38.6% 100.0% 30.3% 39.8%
Atkins 49.0% 71.2% 58.2% 59.5% 68.4% 48.1% 100.0% 54.4%
Graham 51.3% 39.2% 45.5% 41.7% 42.9% 46.0% 39.6% 100.0%

The strongest relationships seem to be between Atkins and Caudill. About two-thirds of voters who chose the one also chose the other.

2014 Boyle County Exit Poll: additional trends in Danville Mayoral election

Here are additional cross-tabulations of voting patterns in the Danville Mayoral race from the 2014 Boyle County Exit Poll. Cross-tabulations with partisanship, city commission approval, Mayor Hunstad approval, Fairness Ordinance approval, and age are found in the previous post here.

“What is the SINGLE most important problem our local area (Boyle county/Danville) needs to solve?

  Stevens Perros  
Crime 42.90% 57.10% 100.0%
Education 62.50% 37.50% 100.0%
Jobs 51.50% 48.50% 100.0%
Economic development 62.20% 37.80% 100.0%
Prescription drug abuse 70.10% 29.90% 100.0%

Looking at it another way, of those who voted for Perros, 31.7% said that “jobs” was the single most important problem facing our local area, followed by 26.2% who said it was “crime.” Those who voted for Stevens similarly claimed “jobs” as the most important issue (28.1%) but were about 10% less likely to report “crime” as the most important problem (16.5%).

Male or female?

  Stevens Perros  
Women 57.30% 42.70% 100.0%
Men 51.70% 48.30% 100.0%

Note: several people declined to answer this particular question on the survey, which skews somewhat the patterns that we observe here which show Stevens winning a clear majority of both groups.

Political ideology:

  Stevens Perros  
Liberal 71.10% 28.90% 100.0%
Moderate 55.90% 44.10% 100.0%
Conservative 40.20% 59.80% 100.0%

Education:

  Stevens Perros  
High school 51.30% 48.70% 100.0%
College graduate 48.30% 51.70% 100.0%
Postgraduate 64.10% 35.90%  

There were no statistically significant differences on family income or frequency of church attendance in predicting voting for Perros vs. Stevens.

Summary:

In general, the two mayoral candidates drew their support from different constituencies. Mike Perros supporters were more likely to be slightly older, conservative, Republican men who are a little more worried more about jobs and crime and who were less supportive of the current city commission (of whom Stevens was a member) and also more opposed to recent important decisions like passing the Fairness Ordinance.

Paige Stevens, on the other hand, drew support from slightly younger, more liberal, Democratic women who were a little more concerned about jobs, education, economic development, and prescription drug abuse. She also tended to be supported by those who were more supportive of the current city commission and by those who supported the passage of the Fairness ordinance. (That being said, it should be emphasized that these are only general trends and patterns and not absolute 1-to-1 relationships.)

Given these patterns, it may be the case that Mike Perros’s victory may be at least partially attributable to the structural advantage of having Danville Mayoral elections coinciding with national midterm elections every four years (2014, 2010, 2006, etc.). In general, midterm elections tend to see lower levels of turnout than national presidential elections, and those with higher socioeconomic levels (in terms of income, education, race, etc.) tend to represent a larger share of the electorate in midterm elections. Thus, the same factors that give Republicans a demographic advantage in national midterm elections may also have advantaged Mike Perros to a minor or moderate extent as he was more likely to receive support from more traditionally conservative demographic groups.

2014 Boyle County Exit Poll: Preliminary Results

The 2014 Boyle County Exit Poll was jointly administered by Benjamin Knoll of Centre College and Ryan New of Boyle County High School. The survey sample consisted of Boyle County voters on November 4, 2014. Respondents were randomly selected by interviewers to participate in the survey. Centre College students enrolled in Professor Benjamin Knoll’s POL 205 (Introduction to Political Analysis), POL 210 (Introduction to American Politics), and POL 330 (Parties, Campaigns, and Elections) courses participated in designing and administering the exit poll surveys. They were joined by students from Boyle County High School’s AP U.S. Government course in administering the surveys.

Students were on-site from 6:00 AM through 6:00 PM surveying voters as they left the polling locations on Tuesday, November 4th, 2014. In all, 1,684 Boyle County voters (including 1,141 self-reported from Danville) participated in the exit poll. The Kentucky Secretary of State’s website reports that 9,720 individuals voted in Boyle County on Election Day. The following statistics therefore have a margin of error of ±2.2% for questions given to all Boyle County voters.

It should be noted that this is an exit poll of voters only. In this election, only about 42% of all adults in Boyle County voted. Therefore, these figures should not be interpreted as fully representative of all adults in Boyle County, but rather less than half of all adults. Given the sample size, however, these figures should be considered representative of all adults who voted on Election Day in 2014.

Figures presented here are also statistically weighted by gender, race, and age.

ALL BOYLE COUNTY VOTERS

Generally speaking, do you believe Danville/Boyle County is heading in the right direction or heading off on the wrong track?

  • 73.6% heading in the right direction, 21% off on the wrong track, 5.3% DK/no opinion

What is the single most important problem that our local area (Boyle County/Danville) needs to solve?

  • 25% Jobs
  • 19.8% Crime
  • 11.9% Prescription drug abuse
  • 11.1% Economic development
  • 9.1% Education
  • 12% DK/no opinion

Do you approve of the way that the following political leaders are handling their job?

  • President Obama: 36.4% approve, 59% disapprove, 4.6% DK/no opinion
  • Senator Rand Paul: 45.8% approve, 41.1% disapprove, 13.1% DK/no opinion
  • Judge Harold McKinney: 67.1% approve, 15.5% disapprove, 17.4% DK/no opinion
    • Of those who approve of Judge McKinney, 78% also voted for him and 22% voted for challenger Lynn Harmon. (excludes DK/no opinion)
    • Of those who disapprove of Judge McKinney, 10% voted for him and 90% voted for challenger Lynn Harmon. (excludes DK/no opinion)

Some of the strongest predictors of voting for Harold McKinney vs. Lynn Harmon for Boyle Judge Executive are Democratic partisanship, levels of education, and whether the voter felt that Danville/Boyle County is on the right track or wrong track:

Harmon (R) McKinney (D)
Democrat 14.4% 85.6% 100%
Independent 37.3% 62.7% 100%
Republican 64.4% 35.6% 100%
  Harmon (R) McKinney (D)
High School 48.3% 51.7% 100%
College Graduate 43.1% 56.9% 100%
Postgraduate 21.5% 78.5% 100%
  Harmon (R) McKinney (D) TOTAL
Right direction 33.7% 66.3% 100%
Wrong direction 52.1% 47.9% 100%

Would you prefer that Rand Paul:

  • Run for president in 2016: 22.5%
  • Run for reelection to Senate in 2016: 46.8%
  • DK/no opinion: 30.8%
  • AMONG REPUBLICANS ONLY: 33.5% run for president, 53% run for reelection to Senate
  • AMONG DEMOCRATS ONLY: 12.8% run for president, 35.1% run for reelection to Senate

Generally speaking, do you think students in your (Danville/Boyle) school district perform:

  • 52.1% better than state average, 34.7% about the same as state average, 6.7% below state average, 6.3% DK/no opinion
  • DANVILLE VOTERS ONLY: 52% better than state average, 39.5% about the same as state average, 8.3% below state average (excludes DK/no opinion)
  • COUNTY VOTERS ONLY: 66.5% better than state average, 29.8% about the same as state average, 3.5% below state average (excludes DK/no opinion)
  • PARENT/GUARDIAN OF STUDENT ONLY: 61.1% better than average, 31.5% about the same as state average, 7.4% below state average (excludes DK/no opinion)

DANVILLE VOTERS ONLY

Do you approve of the way that the following political leaders are handling their job?

  • Danville Mayor Bernie Hunstad: 29.1% approve, 55% disapprove, 15.9% DK/no opinion
  • Danville City Commission: 52.2% approve, 28.3% disapprove, 19.2% DK/no opinion

Do you approve or disapprove of Danville City Commission’s handling of the following issues:

  • Fostering economic growth: 50.4% approve, 32.7% disapprove, 16.8% DK/no opinion
  • Fostering a good quality of life for residents: 61.5% approve, 21.1% disapprove, 17.2% DK/no opinion
  • Management of the city budget: 48.9% approve, 31.7% disapprove, 19.2% DK/no opinion
  • Management of the water plant project: 58.5% approve, 21% disapprove, 20% DK/no opinion

Regarding Danville’s new “Fairness Ordinance,” do you:

  • 29.4% strongly approve, 26.2% somewhat approve, 13.9% somewhat disapprove, 17.8% strongly disapprove, 12.6% DK/no opinion (63.7% approve, 36.3% disapprove when excluding DK/no opinion)
    • DEMOCRATS ONLY: 82.5% approve (excludes DK/no opinion)
    • REPUBLICANS ONLY: 39.1% approve (excludes DK/no opinion)
    • INDEPENDENTS ONLY: 62.2% approve (excludes DK/no opinion)
    • FREQUENT CHURCH ATTENDERS ONLY: 53.4% approve (excludes DK/no opinion)
    • CHURCH NON-ATTENDERS ONLY: 80.4% approve (excludes DK/no opinion)
    • EVANGELICAL PROTESTANT ONLY: 48.8% approve (excludes DK/no opinion)
    • MAINLINE PROTESTANT ONLY: 68.6% approve (excludes DK/no opinion)
    • CATHOLIC: 67.8% approve (excludes DK/no opinion)
    • RELIGIOUSLY NON-AFFILIATED ONLY: 71.8% approve (excludes DK/no opinion)

Mike Perros beat sitting Commissioner Paige Stevens 51%-49%. Some of the strongest predictors of voting for Mayor-Elect Mike Perros are Republican partisanship, disapproval of the current Danville City Commission, approval of Mayor Hunstad, and disapproval of the Danville Fairness Ordinance, and age:

  Stevens Perros TOTAL
Democrat 64.8% 35.2% 100%
Independent 64.8% 35.2% 100%
Republican 42.1% 57.9% 100%
  Stevens Perros TOTAL
Approve of City Commission 64.2% 35.8% 100%
Disapprove of City Commission 39.0% 61.0% 100%
  Stevens Perros TOTAL
Approve of Mayor Hunstad 43.0% 57.0% 100%
Disapprove of Mayor Hunstad 61.1% 38.9% 100%
  Stevens Perros TOTAL
Approve of Fairness Ordinance 64.7% 35.3% 100%
Disapprove of Fairness Ordinance 39.1% 60.9% 100%
Stevens Perros Total
Under 35 63.9% 36.1% 100.0%
35-55 61.1% 38.9% 100.0%
55-75 53.0% 47.0% 100.0%
Over 75 47.0% 53.0% 100.0%

There were also some interesting patterns in voting for Danville City Commission members:

  • Those who voted for J.H. Atkins were also moderately likely to vote for Kevin Caudill and against Kent Mann and Buck Graham. They’re also fairly approving of the City Commission’s handling of various issues and approving of the Fairness Ordinance. They’re also likely to be Democrats and have more education.
  • Kevin Caudill drew support from similar groups as J.H. Atkins, except there’s no clear partisan support one way or another. He also draws more support from those with higher incomes more education.
  • Those who voted for Buck Graham were likely to moderately disapprove of the City Commission’s handling of various issues and to disapprove of the Fairness ordinance. He also drew some support from older voters and Republican voters.
  • Those who voted for Rick Serres were slightly more likely to also vote for Denise Terry but against J.H. Atkins. He also drew some support from more conservative voters and those with higher incomes and more education.
  • Those who voted for Denise Terry and Kent Mann were also likely to have slightly higher approval ratings of Mayor Hunstad but slightly more likely to disapprove of the Commission as a whole.

DEMOCRAPHIC INFORMATION OF ALL SURVEY RESPONDENTS (added 11/5/2014):

Do you consider yourself: Liberal: 22.7%, Moderate: 36.9%, Conservative: 37.4%, 2.9% DK/no opinion

Do you think of yourself as a (an): Democrat/leaner: 47.3%, Independent: 5.3%, Republican/leaner: 43.2%, 4.2% DK/no opinion

Additionally, 53.4% of survey respondents were female, 46.5% male; 61% report an income over $50K/year while 9.9% report an income under $20K/year; 13% report never attending church, 29% report attending sometimes, and 55.1% report attending once a week or more; 10.3% report being a “Tea Party supporter”; 93% report white ethnicity with 5.1% reporting African-American ethnicity; 32.6% report a high school education or less, 36.9% report college level of education, and 29.1% report a post-graduate level of education. 10.7% reported being under 35 years old, 31% reported being between 35-55 years old, 40.3% reported being between 55-75 years old, with 17.7% report being older than 75. Evangelical Protestants make up 38% of the sample, with 27.5% Mainline Protestants, 9.9% Catholic, 8.6% religiously unaffiliated, 5.7% “other” religion.

2014 BCEP results to be posted

This year dozens of students from Centre College are teaming up with Boyle County High School’s AP US Government course to field the surveys for the 2014 Boyle County Exit Poll.

The topline results from the Exit Poll will be posted here late Tuesday/early Wednesday and follow-up analyses will be posted throughout Wednesday and the rest of the week.

Updates will also be posted to my Twitter feed: @benjaminknoll28