North Dakota Statewide Survey, October 2014 #785-AJND

Prepared by DFM Research

1 | P a g e

Methodology

The survey results presented in this report are based on a stratified random sample of 430 North
Dakotans that indicated that they are certain or very likely to vote in the 2014 general election.
The sample was stratified by region to ensure a representative sample of the North Dakota
electorate; the stratified sample was comprised of five distinct regions:

26% - Eastern Cities (Fargo, West Fargo, Grand Forks)
20% - Western Cities (Bismarck, Mandan, Minot)
18% - Eastern North Dakota
19% - Central North Dakota
17% - Western North Dakota

The percentages allocated for each region was based on voter turnout in the 53 counties from the
2010 and 2012 general election, coupled with the most recent Census Bureau state population
estimates.

After the numbers were stratified into the appropriate region, telephone numbers were then selected
by random using a skip pattern to guarantee that the interviews were distributed throughout the
region. Each number in the stratified sample had the same non-zero chance of being selected for an
interview.

Telephone interviews were conducted by trained staff of Stone Research Services of Indianapolis,
Indiana, using a computer-assisted telephone interview (CATI) system for landline phones. Cell
phone interviews are dialed manually to comply with the Telemarketing Consumer Protection Act
of 1991. To ensure everyone in the household would have an equal chance of being selected; callers
would ask to interview the resident over the age of 18 who had the most recent birthday.

Final results are weighted based on gender, age, and education to conform to the approximate
voting population based on U.S. Census Bureau demographic data and election statistics. Below are
the final demographic percentage used to weight the survey.

Gender Age Education
Male 49% 18-39 30% High School/Less 29%
Female 51% 40-64 46% Some College/AA 38%
65+ 24% Bachelor/Graduate 33%

The final results presented are subject to sampling error, which is the difference between results
obtained from the survey, and those if everyone in the target population were interviewed. The
sampling error, commonly known as the margin of error, is + 4.7 percentage points with a 95
percent confidence level; meaning that in 19 out of 20 times, the individual responses would be
within the margin of error (confidence interval). If final results of a question resulted in a tabulated
answer of 50 percent, the confidence interval would between 45.3 to 54.7 percent. Where
appropriate, question and answer choices are randomized to reduce order bias; due to rounding,
numbers may not equal 100 percent.

Project management and final analysis of the data was completed by Dean Mitchell of DFM
Research based in Saint Paul, Minnesota. In addition to his 22 years of political experience, Dean
has completed course work in survey techniques and statistics as part of his Master in Public Policy
(MPP) degree from the University of Minnesotas Humphrey School of Public Affairs.

North Dakota Statewide Survey, October 2014 #785-AJND
Prepared by DFM Research
8 | P a g e

Q6: If the election was held today for Secretary of State, would you vote for April Fairfield the
Democrat, Al Jaeger the Republican, or Roland Riemers the Libertarian?

Al Jaeger ........................................................................................... 46%
April Fairfield ................................................................................... 26
Roland Riemers ................................................................................ 4
(VOL) Unsure ................................................................................... 24

Gender Jaeger Fairfield Riemers Unsure

Men 49% 24 7 20
Women 43 28 2 28

Age Jaeger Fairfield Riemers Unsure

18-39 40 26 9 25
40-64 51 24 2 23
Over 65 44 29 3 24

Education Jaeger Fairfield Riemers Unsure

High School/Less 41 27 4 28
Some College/AA Degree 44 24 6 26
Bachelor/Graduate Degree 53 27 2 18

Region Jaeger Fairfield Riemers Unsure

East City 45 23 5 28
West City 44 32 6 18
East Rural 38 31 5 26
Central Rural 52 27 1 20
West Rural 52 17 5 27

Party Identification Jaeger Fairfield Riemers Unsure

Democrat 14 67 0 19
Independent 34 23 7 36
Republican 80 3 3 14

North Dakota Statewide Survey, October 2014 #785-AJND
Prepared by DFM Research
9 | P a g e

Q7: If the election was held today for Attorney General, would you vote for Kiara Kraus-Parr the
Democrat or Wayne Stenehjem the Republican?

Wayne Stenehjem ............................................................................. 68%
Kiara Kraus-Parr ............................................................................... 17
(VOL) Unsure ................................................................................... 14


Gender Stenehjem Parr Unsure

Men 74% 14 11
Women 62 20 17

Age Stenehjem Parr Unsure

18-39 68 20 13
40-64 69 16 15
Over 65 66 18 15

Education Stenehjem Parr Unsure

High School/Less 59 25 16
Some College/AA Degree 65 18 17
Bachelor/Graduate Degree 80 10 10

Region Stenehjem Parr Unsure

East City 70 19 12
West City 65 23 12
East Rural 63 16 22
Central Rural 66 17 17
West Rural 79 11 11

Party Identification Stenehjem Parr Unsure

Democrat 37 50 13
Independent 66 11 23
Republican 91 3 5