[Last revised 9/12/06

 

 

BEST IF VIEWED WITH INTERNET EXPLORER

 

NOTE: DUE TO GRAPHICS SOME PAGES MAY LOAD SLOWLY

 

 

 

St. Cloud State University SURVEY

  STATEWIDE SURVEY OF MINNESOTA ADULTS

2004

 

 

 

* LINK TO COMPUTER ASSISTED QUESTIONNAIRE

* LINK TO REPORTS

 

Dr. Stephen Frank

Dr. Steven Wagner

Dr. Michelle Kukoleca Hammes

 

Principal Investigateurs

SCSU Survey

 

Social Science Research Institute

College of Social Sciences

St. Cloud State University

 St. Cloud, Minnesota

 

 

Drs. Frank, Wagner and Kukoleca are members of the Midwest Association of Public Opinion Research (MAPOR) and the American Association of Public Opinion Research (AAPOR) and subscribe to the code of ethics of the AAPOR.

METHODOLOGY, MISSION, BACKGROUND

 

I. History and Mission of the Survey

 

The SCSU Survey is an ongoing survey research extension of the Social Science Research Institute in the College of Social Science at St. Cloud State University.  The SCSU Survey performs its research in the form of telephone interviews.  Telephone surveys are but one of the many types of research employed by researchers to collect data randomly.  The telephone survey is now the instrument of choice for a growing number of researchers.

 

Dr. Steve Frank began the SCSU Survey in 1980 conducting several omnibus surveys a year of central Minnesota adults in conjunction with his Political Science classes. The omnibus surveys are now done once a year. In addition to questions focusing on the research of the faculty directors, clients can buy into the survey or contract for specialized surveys.

 

Presently, the omnibus surveys have continued, but have shifted to a primary statewide focus.  These statewide surveys are conducted once a year in the fall and focus on statewide issues such as election races, current events, and other important issues that are present in the state of Minnesota.  Besides the annual fall survey, the SCSU Survey conducts an annual spring survey of SCSU students on various issues such as campus safety, alcohol and drug use, race, etc.  Lastly, the SCSU Survey conducts contract surveys for various public and private sector clients.  The Survey provides a useful service for the people and institutions of the State of Minnesota by furnishing valid data of the opinions, behaviors, and characteristics of adult Minnesotans.

 

The primary mission of the SCSU Survey is to serve the academic community and various clients through its commitment to high quality survey research and to provide education and experiential opportunities to researchers and students.  We strive to assure that all SCSU students and faculty directors contribute to the research process, as all are essential in making a research project successful.  This success is measured by our ability to obtain high quality survey data that is timely, accurate, and reliable while maintaining an environment that promotes the professional and personal growth of each staff member.  The survey procedures used by the SCSU Survey adhere to the highest quality academic standards.  The SCSU Survey maintains the highest ethical standards in its procedures and methods.  Both faculty and student directors demonstrate integrity and respect for dignity in all interactions with colleagues, clients, researchers, and survey participants.

 

II. Survey Staff

 

The Survey’s faculty directors are Dr. Steve Frank (SCSU Professor of Political Science), Dr. Steven Wagner (SCSU Associate Professor of Public and Non-Profit Administration) and Dr. Michelle Kukoleca Hammes (SCSU Assistant Professor of Political Science).  The faculty directors are members of the Midwest Association of Public Opinion Research (M.A.P.O.R.) and the American Association of Public Opinion Research (A.A.P.O.R.). The directors subscribe to the code of ethics of A.A.P.O.R.

 

Stephen I. Frank

 

Dr. Frank holds a Doctor of Philosophy in Political Science from Washington State University.  Dr. Frank teaches courses in American Politics, Public Opinion and Research Methods at St. Cloud State University.  Dr. Frank started the SCSU Survey in 1980 and has played a major role in the development, administration and analysis of over 150 telephone surveys for local and state governments, school districts and a variety of nonprofit agencies.  Dr. Frank has completed extensive postgraduate work in survey research at the University of Michigan.  Dr. Wagner and Dr. Frank have published two texts on Minnesota’s former Governor, Jesse Ventura. Frank and Wagner’s newest publication is The Maverick Campaign and Election of Jesse Ventura—S.C. Wagner and S.I. Frank in Campaigns and Elections: Issues, Concepts, Cases Robert P. Watson and Colton C. Campbell, editors 3/2003.  Frank has recently had published “New Directions In Public Opinion” in Perspectives On Minnesota Government & Politics 5th Ed.  Steven Hoffman, Homer Williamson and Kay Wolsborn editors, June 2003.  Frank is currently serving as President of the Minnesota Political Science Association. 

 

Steven C. Wagner

 

Dr. Wagner holds a Doctor of Philosophy in Political Science and a Master of Public Administration from Northern Illinois University.  Dr. Wagner earned his Bachelor of Science in Political Science from Illinois State University.  Dr. Wagner teaches courses in American Politics and Public and Nonprofit Management at St. Cloud State University.  Dr. Wagner joined the SCSU Survey in 1997.  Before coming to SCSU, Dr. Wagner taught in Kansas where he engaged in community-based survey research and before that was staff researcher for the U.S. General Accounting Office.  Dr. Wagner has written many papers on taxation, health care delivery and state politics and has published articles on voting behavior, federal funding of local services and organizational decision-making. Dr. Wagner and Dr. Frank have published two texts on Minnesota’s former Governor, Jesse Ventura. Frank and Wagner’s newest publication is The Maverick Campaign and Election of Jesse Ventura—S.C. Wagner and S.I. Frank in Campaigns and Elections: Issues, Concepts, Cases Robert P. Watson and Colton C. Campbell, editors 3/2003. 

 

Michelle K. Hammes

 

Dr. Kukoleca Hammes holds a Doctor of Philosophy in Political Science and a Masters in Political Science from the State University of New York at Binghamton.  Dr. Kukoleca Hammes earned her Bachelor of Arts in Political Science from Niagara University.  Dr. Kukoleca Hammes’ is a comparativist with an area focus on North America and Western Europe.  Her substantive focus is representative governmental institutions.  She teaches courses in American Government, Introduction to Ideas and Institutions, Western European Politics, and a Capstone in Political Science at St. Cloud State University.  Dr. Kukoleca Hammes has recently joined the survey team and will be using her extensive graduate school training in political methodology to aid in questionnaire construction and results analysis.  Kukoleca Hammes has recently had published “The ‘State’ of Participation” in Perspectives On Minnesota Government & Politics 5th Ed.  Steven Hoffman, Homer Williamson and Kay Wolsborn editors, June 2003.  Kukoleca Hammes is currently serving on the board of the Minnesota Political Science Association.

 

STUDENT DIRECTORS AND TECHNICAL STAFF

  

STUDENT SUPERVISING DIRECTOR

Mr. Jason Lunser, Senior, Political Science Major, International Relations and Economics Minors, Cold Spring, Minnesota

 

SCSU Survey Lab Student DIRECTORS/Consultants

Mr. Chris Brixius, Junior, Sociology Major, St. Cloud, Minnesota

Ms. Adriana Dobrzycka, Senior, Anthropology and Political Science Majors, Spanish Minor, Florence, Italy

Mr. Michael Fox, Senior, Political Science Major, International Relations Minor, Breckenridge, Minnesota

Ms. Nicole Kahler, 4th Year Student, Social Work Major, Roseville, Minnesota

Ms. Sara Lohrman, Sophomore, Political Science Major, Willmar, Minnesota

Mr. Joshua Mattison, 4th Year Student, Political Science Major, Sociology Minor, Thief River Falls, Minnesota

Ms. Stacey Springer, Senior, Psychology and Political Science Majors, Lincoln, Nebraska

 

                                                         

student Technical Consultant

Mr. Jason Amunrud, Junior, Computer Science Major, Shoreview, Minnesota

 

After five or more hours of training and screening, approximately 50 students from Political Science 101 and 195 classes (introductory Political Science and introductory American National Government) taught by Drs. Frank and Hammes completed the calling.  Faculty directors monitored the calling shifts.  Student directors conducted both general training sessions and one-on-one training sessions as well as monitoring all calling shifts.

 

III. Methodology

 

The SCSU Survey is operated out of Stewart Hall 324.  It is also known as the CATI Lab, which stands for Computer Assisted Telephone Interviewing Lab.  It is equipped with 13 interviewer stations that each includes a computer, a phone, and a headset.  In addition to the interviewer stations, there is the Supervisor Station, which is used to monitor the survey while it is in progress. The SCSU Survey has its own server designated solely for the use of the survey. 

 

The SCSU Survey is licensed to use Sawtooth Software’s Ci3 Questionnaire Authoring Version 4.1, a state-of-the-art windows-based computer-assisted interviewing package.  This program allow us to develop virtually any type of questionnaire while at the same time programming edit and consistency checks and other quality control measures to insure the most valid data.  Interviewing with Ci3 offers many advantages:

 

1.              Complete control of what the interviewer sees;

2.              Automatic skip or branch patterns based on previous answers, combinations of answers, or even mathematical computations performed on answers;

3.              Randomization of response categories or question order;

4.              Customized questionnaires using respondents’ previous responses, and,

5.              Incorporation of data from the sample directly into the sample database.

6.              All interview stations are networked for complete, ongoing sample management.

7.              Data is updated immediately, ensuring maximum data integrity and allowing clients to get progress reports anytime.  Data is reviewed for quality and consistency.

8.              Answers are entered directly into the computer.  Keypunching is eliminated, thus decreasing human error.  Data analysis can start immediately.

9.              The computer handles call record keeping automatically, allowing interviewers and supervisors to focus on the interviewing task.

10.          Callbacks are handled by the computer and made on a schedule.  We call each number ten times.  Interrupted surveys are easily completed.  Persons who are willing to be interviewed can do so when it is convenient to them, improving the quality of their responses.

11.          Calls are made at various times during the week (Monday through Thursday, 4:30 to 9:30) and on weekends (Sunday, 2:30 to 9:30) to maximize contacts and ensure equal opportunities to respond among various demographic groups. Some daytime calls were made

12.          Some calls were made to Spanish speaking respondents.

13.          CATI maintains full and detailed records, including the number of attempts made to each number and the disposition of each attempt.

 

The survey was administered from Sunday, October 17 through Tuesday, October 26, not on Fridays or Saturdays.  Most calls were made after 4:30 PM weekdays and also during the afternoon on the Sunday calling dates.

 

Several steps were taken to ensure that the telephone sample of Minnesota adults who were eighteen years of age or older was representative of the larger population. Survey Sampling Inc. of Fairfield, Connecticut prepared the random digit sample of telephone numbers. Random digit dialing makes available changed, new, and unlisted numbers. Drawing numbers from a telephone book may skip as many as 20 percent of Minnesota households. Within each household the particular respondent was determined in a statistically unbiased fashion. This means that the selection process alternated between men and women and older and younger respondents. Few substitutions were allowed. In order to reach hard-to-get respondents each number was called up to ten times over different days and times and appointments made as necessary to interview the designated respondent at her/his convenience.

 

We have found Survey Sampling a particularly efficient sample production company.  They generate samples of very high quality because they:

 

1.              construct a comprehensive database of all telephone working blocks which actually represent residential telephones;

2.              obtain, update and cross check working block information from the local (U.S. West) telephone company;

3.              confirm the estimated number of residential telephones with each working block, excluding sparsely populated working blocks (industry standard is to exclude those blocks with less than three known working residential telephones out of the 100 possible numbers);

4.              assign working blocks known to contain residential telephones to geographic areas bases on zip code and most recent updates of census data;

5.              mark each working block for demographic targeting;

6.              check each RDD number against a list of known business telephone numbers and generate new numbers as necessary; and,

7.              arrange the ending sample in a random order to eliminate potential calling order bias.

 

In samples of 673 interviews the overall sample error due to sampling and other random effects is less than 4 percentage points at the 95 percent confidence level. This means that if one were to have drawn 20 samples of the state and administered the same instrument it would be expected that the overall findings would be greater/lesser than 4 percent only one time in twenty.  However, in all sample surveys there are other possible sources of error for which precise estimates cannot be calculated. These include interviewer and coder error, respondent misinterpretation, and analysis errors. When analysis is made of sub-samples such as respondents who are Republicans, or when the sample is broken down by variables such as gender the sample error may be larger.

 

The demographics of the sample match census and other known characteristics of the larger state population very well.  Usually surveys have to employ a statistical technique called weighting on demographics such as sex.  Most surveys usually over-sample females. The ratio of male to female adults in the sample was 48 to 58 percent, which almost matches the actual adult population. Although not needed the sample was weighted for sex. Other variables such as household income, political party affiliation and employment all closely match what is known of the Minnesota adult population.

 

The cooperation rate of the survey was 79 percent. This is above the average for professional marketing firms. When the SCSU Survey conducts specialized contract surveys, we use a smaller, more skilled group of student interviewers and the completion rate ranges often approach 80+ percent. Cooperation rate means that once an eligible household was reached, almost six of ten respondents agreed to participate in the survey.

 

The total survey consisted of 46 variables. Additional information was generated from the sample for area codes and country. Additional material on the survey's methodology and findings are available by contacting Steve Frank, Steven Wagner, or Michelle Kukoleca Hammes.  Contact information can be found on the back page of this report.

 

 

 

Table 1: Calling Record

 

Disposition Record

Frequency

Completed Calls

673

Not Working Numbers

887

Not Eligible - Respondent not available during the period of the study, language problems, hearing problems, not a Minnesota resident, cabin phone, illness, etc.

105

Callbacks - Appointments made but contact could not be made with designated respondent.

426

Refusals - Attempt to re-contact and convert refusals to a completion was made for most refusals.

177

Answering Machine - Live contact could not be made even after nine calls.

364

Business Phone

302

No Answers - Probable non-working numbers but some may be households on vacation, etc.

408

Fax/Modem

107

Busy

68

Call Blocking

6

Partial - Complete except for demographics

2

Partial - Incomplete, more than demographics left.

14

Total Calls Placed

3544

 

 

IV. Demographics

 

Methodological Notes

 

Shown below are frequency tables of the demographic indicators we collected as part of the sample or asked of the respondents.  Also, we show demographic tables of party, age, income, and employment with some categories combined to facilitate cross tabulation analysis.  The tables labeled “recoded” are used in the cross tabulation analysis.

 

 

 

Table 2: Gender

 

RESPONSE

FREQUENCY

PERCENT

Male

329

49

Female

344

51

Total

673

100%

 

 

 

Table 3: Party Identification

 

RESPONSE

FREQUENCY

PERCENT

Always Votes Democratic

110

16

Democrat Who Sometimes Votes for Other Party

114

17

Always Votes Republican

68

10

Republican Who Sometimes Votes for Other Party

94

14

Always Votes Green

2

0

Green Who Sometimes Votes for Other Party

6

1

Always Votes MN Independence

5

1

MN Independence Who Sometimes Votes for Other Party

23

3

independent Closer to Democrats

52

8

independent Closer to Republicans

54

8

independent Closer to Green

4

0

independent Closer to MN Independence Party

52

8

Other

38

6

Apolitical

18

3

Don’t Know/ Refused

33

5

Total

673

100%

 

 

 

Table 4: Recoded Party

 

RESPONSE

FREQUENCY

PERCENT

Democrat

276

41

Republican

216

32

Green

12

2

Independence

80

12

Other

38

5

Don’t Know/ Missing

33

5

Total

673

10%

 

 

 

Table 5: Age

 

RESPONSE

FREQUENCY

PERCENT

18-24

63

9

25-34

81

12

35-44

121

18

45-54

159

24

55-65

118

17

65+

125

19

Don’t Know/ Refused

6

1

Total

673

100%

 

 

 

Table 6: Recoded Age

 

RESPONSE

FREQUENCY

PERCENT

18-34

144

21

35-64

398

59

65+

125

19

Don’t Know/ Missing

6

1

Total

673

100%

 

 

 

Table 7: Employment

 

RESPONSE

FREQUENCY

PERCENT

Working Now

407

61

Laid Off

12

2

Unemployed

14

2

Retired

162

25

Disabled

17

3

Household Manager

29

4

Student

22

3

Don’t Know/ Refused

1

0

Total

673

100%

 

 

 

Table 8: Combined Household Income Level

 

RESPONSE

FREQUENCY

PERCENT

Under $10,000

20

3

$10,001-$15,000

24

3

$15,001-$20,000

20

3

$20,001-$25,000

21

3

$25,0001-$30,000

39

6

$30,001-$40,000

54

8

$40,001-$50,000

55

8

$50,001-$100,000

107

16

$100,000+

151

22

Refused

44

7

Don’t Know

139

21

Total

673

100%

 

 

Table 9: Recoded Income Level

 

RESPONSE

FREQUENCY

PERCENT

Under $25,000

85

12

$25,001-$50,000

148

23

$50,001-$100,000

107

16

$100,000+

151

22

Don’t Know/ Refused/ Missing

182

27

Total

673

100%

 

 

Table 10: Area Code

 

RESPONSE

FREQUENCY

PERCENT

218

90

13

320

67

10

507

121

18

612

47

7

651

150

22

763

98

15

952

100

15

Total

673

100%

 

 

Table 11: County Code from Sample

 

RESPONSE

FREQUENCY

PERCENT

Seven Metro Counties

363

54

Greater Minnesota Counties

310

46

Total

673

100%