HIGHLIGHTS

• Ninety-five percent of Minnesota adults have gambled at sometime during their lives.

• Eighty-three percent gambled in the year prior to the April 2003 survey.

• Although participation from activity to activity varies, the percent of Minnesota adults who choose to wager has remained stable since this survey began in 1993.

• The average Minnesota adult has wagered on six gambling activities in their lifetime, and three in the past year.

• In 2003, more people played the Minnesota State Lottery than any other form of gambling in the state.

• Indian casinos have experienced the most growth in participation since 1993.

INTRODUCTION

"Hello…I’m calling from St. Cloud State University. We are surveying residents of the state, to determine how residents feel about various issues such as lotteries and gambling. We would like to begin by listing several types of gambling activities. For each one, please tell me if you have ever made a wager on that type of gambling and if you have done it in the past year. Let’s start with social bets with friends or relatives…"

Since 1993, the Minnesota State Lottery and St. Cloud State University have been surveying to gather information on gambling practices and attitudes toward gambling from Minnesota adults.

The Minnesota State Lottery designed this survey in conjunction with the St. Cloud State University Survey. The SCSU Survey designed the sample and conducted the interviews. Lottery staff completed the data analysis and report writing. The survey asked questions regarding participation and perceptions about sixteen forms of gambling available to Minnesota adults, as well as lottery specific questions and general background information.

Seventy-nine percent of the households contacted completed the survey, resulting in 2002 completed interviews.¹

The design, methodology, and many of the survey’s questions were replicated in phone interviews conducted in 1993, and 1995 through 2002. This report uses data from earlier surveys to determine whether gambling behavior in Minnesota adults has changed in recent years.

¹With a sample of this size, the estimated sample error statewide is approximately plus or minus 2.20 percent at the 95 percent level of confidence; if one were to draw repeated samples of the same size and administer the same survey, the findings would differ by more than 2.20 percent only one time in 20. Sample errors for demographic or other subgroups may be larger.

LIFETIME GAMBLING PARTICIPATION

Ninety-five percent of those responding to the survey reported participating in at least one of the 16 forms of gambling available to Minnesota adults sometime during their lives- that’s about 3.4 million of the 3.6 million adults living in the state.

The lifetime gambling participation rose by 8 percent between 1993 and 2003, from 87 percent to 95 percent. With minor exceptions, the same gambling activities, questions, and sampling approach were used each respective year.² If only the original gambling activities from 1993 were sampled in 2003, the lifetime participation rate would have risen from 87% to 93%.

• Lifetime Gambling Participation CHART

² The measure in 1995 inadvertently left out the question on sports betting. Questions about raffles and internet gambling have been added since 1993 and are included in this tabulation. In 2002, we split the question about non-Indian casinos into three parts – riverboats, cruise ships, and other casinos.

NUMBER OF LIFETIME GAMBLING ACTIVITIES

According to the 2003 data the average Minnesota adult has wagered on six of the 16 gambling activities. This number has increased since the beginning of the survey in 1993, when the average adult had only wagered on four of the activities. (This increase can be partially attributed to the gambling activities added to the survey since 1993). No one reported trying them all.

• Number of Lifetime Gambling Activites CHART

PARTICIPATION IN FORMS OF GAMBLING

The most popular form of gambling among Minnesotans is raffles; 77 percent of Minnesota adults report buying a raffle ticket at some point in their lives. Close behind is the Minnesota State Lottery (69 percent) and Indian casinos (64 percent). The lowest gambling participation rates are found for dice, dog racing, and the Internet.

• "Have you ever bet on..." CHART

GAMBLING PARTICIPATION IN THE PAST YEAR

Gambling has been a part of most Minnesota adults’ lives at some point, but how many gambled during the year prior to the April 2003 survey?

Eighty-three percent (about 3.0 million adults) reported engaging in some form of gambling during the year prior to the survey. The other 17 percent of the adult population (18 and older) reported no wagers during this year. This participation rate has remained relatively stable throughout the years of the survey.

• Gambling Participation in the Past Year CHART

NUMBER OF PAST YEAR GAMBLING ACTIVITIES

According to the 2003 data the average Minnesotan wagered on three of the 16 gambling activities during the year preceeding the survey. As seen in the past year participation rate, this number has also remained relatively stable since the beginning of the survey. No one reported trying them all.

• Number of Past Year Gambling Activities CHART

PARTICIPATION IN FORMS OF GAMBLING

In the year prior to the April 2003 survey the most popular form of gambling for a Minnesota adult was the Minnesota State Lottery. Fifty-nine percent of the eligible population, or about 2.1 million Minnesotans, bought a lottery ticket. Close behind in popularity are raffles (47 percent), and Indian casinos (40 percent). Riverboats/cruises, horse and dog racing, and the Internet had the lowest participation rates. The strong popularity of the Lottery and raffles and the lack of popularity of racing and riverboats is consistent with previous surveys.

• Have you made this kind of bet within the past year? CHART

GAMBLING THROUGHOUT THE YEARS

THE CHANGING FACE OF GAMBLING

With overall gambling participation rising slightly each year, the specific games of interest for Minnesota adults have also changed. The following chart illustrates each forms gain or loss since 1995. The charts on page 12 compares the lifetime participation rates from years 1995, 1999, and 2003 to illustrate the rate of how gambling interests have changed.³ Indian casinos have experienced the highest rate of overall growth, while cards and other (non-Minnesota) lotteries have shown declines.

• Change in Lifetime Gambling Participation CHART

³ Questions about raffles, internet gambling, riverboats, and cruise ships have been added to the survey but are not included in this chart for the sake of comparability. The question asking about sports bets was accidentally omitted in 1995.

LIFETIME GAMBLING PARTICIPATION

• Have you ever bet on... CHART

PAST YEAR GAMBLING THROUGHOUT THE YEARS

The past year participation rate reflects the slight rise in gambling participation seem in lifetime participation rates. The following chart illustrates change in Past Year participation from 1995 to 2003. The chart on Page 14 compares the past year participation from years 1995, 1999, and 2003 to further illustrate how gambling interests have changed.⁴

⁴ Questions about raffles, internet gambling, riverboats, and cruise ships have been added to the survey but are not included in this chart for the sake of comparability. The question asking about sports bets was accidentally omitted in 1995.

• Change in Past Year Gambling Participation CHART

PAST YEAR GAMBLING PARTICIPATION

• In the past year, have you bet on... CHART

PLAYER RETENTION

When a customer tries a form of gambling, do they return to that activity or is it an experiment to which they do not return? One way of measuring customer retention is to compare the number of people who report taking part in an activity during their lifetime with those who report taking part in the past year. The chart below shows, for example, that 85 percent of those who have purchased a lottery ticket during their lifetime also did so during the past year. The Minnesota lottery, Internet gambling⁵, local casinos and social bets have maintained the greatest percentage of players. Dog racing, horse racing, and gambling on cruise ships have retained the lowest number of participants. This could be due to the seasonal nature of the activity, cost and distance to a cruise, or the closing of the dog track in Hudson, Wisconsin.

⁵Figures on Internet gambling may not be reliable due to the small number of people reporting this activity.

• Player Retention CHART

DEMOGRAPHICS OF GAMBLERS

This section compares gambling participation rates of respondents by age, education, income, and gender to see if differences exist between these subpopulations.

AGE

The following graph shows a comparison of age and gambling participation. Participation peaks in the 25 to 34 year old age group and slowly declines with age.

• Gambling Participation by age CHART

EDUCATION

Educational attainment shows little relationship to gambling participation. Fewer people without a high school diploma report gambling, although the majority still participate. This may be more a function of age since nearly half of this group is composed of those 65 and older.

• Gambling participation by education CHART

HOUSEHOLD INCOME

Past year gambling participation rates are higher beginning with the $25-$35 thousand level of annual household income, then remain consistent though the highest income category. Lifetime participation rates are relatively equal among categories. Income is probably the most difficult category to interpret since many variables hide within it. Size of the household, number of dependents, and number of income producers all affect the income available for gambling and other discretionary purposes.

• Gambling participation by household income CHART

GENDER

Who gambles more – men or women? Overall, men are slightly more likely to gamble than women, with 96% of Minnesota males having bet at sometime during their lives, compared to 94% of females. Raffles, Indian casinos, and cruise ships/riverboats are equal, while men bet at significantly higher rates on cards, social bets, and sports events, and women bet at higher rates on bingo. The chart on Page 20 shows the lifetime participation rates of each gambling activity by gender.

• Participation by gender CHART

METHODOLOGY

During April and May 2003, the Minnesota State Lottery contracted with the St. Cloud State University Survey for a telephone survey of Minnesota adults. The SCSU Survey is an ongoing survey research arm of the Social Science Research Institute in the College of Social Sciences at St. Cloud State University. The questionnaire was designed by the Minnesota State Lottery and modified in consultation with the SCSU Survey directors.

Directors of the Survey are Dr. Steve Frank, SCSU Professor of Political Science, Dr. Steven Wagner, SCSU Associate Professor of Public Administration and Dr. Michelle Kukoleca Hammes, SCSU Assistant Professor of Political Science. Drs. Frank, Wagner, and Kukoleca Hammes are members of the American and Midwest Association of Public Opinion Research and subscribe to the code of ethics of the AAPOR. About 35 trained, supervised, and paid SCSU students conducted the actual interviews using a computer-assisted telephone interviewing (CATI) system. All calls were made from the SCSU Survey Research laboratory. Callers were identified as being from St. Cloud State University. The survey was administered between April 6 and May 5, 2003. The instrument was pre-tested by SCSU Survey prior to the start of the phone calling.

Several steps were taken to ensure that the telephone sample of adult Minnesotans was representative of the larger Minnesota population. Households were contacted using random digit dialing, a technique that can reach changed, new and unlisted phone numbers (samples from phone books may skip 30 percent of more of area households). Survey Sampling, Inc., of Fairfield, CT prepared the random digit sample. Within each household the particular respondent was determined in a statistically unbiased fashion using the Hagen-Collier method.⁶ This method seeks to eliminate statistical bias by alternation between men and women and older and younger respondents. Few substitutions were allowed. Hard-to-reach respondents were called up to 12 times over different days and times, and appointments were made as necessary to interview the designated respondent at his or her convenience. Calls were made after 4:00 p.m., Sunday through Thursday, and calls on weekends and days were made if necessary.

The sample was drawn proportional to the state population. The completed sample consists of 2,002 interviews. In samples of 2,002 interviews, the sample error due to sampling and other random effects is approximately plus/minus 2.20 percent at the 95 percent confidence level. This means that if one were to have drawn 20 samples of the population and administered the same instrument it would be expected that the overall findings would be greater/lesser than 2.20 percent only one time in twenty. As with all sample surveys, however, there are other possible sources of error for which precise estimates cannot be calculated. For sub-samples such as age and gender, the sample error may be larger.

The cooperation rate of the survey is 79 percent. This is about 20 percentage points above the average for professional marketing firms. The cooperation rate means that once an eligible household was reached, almost eight in ten respondents agreed to participate in the survey. A total of 2,002 interviews were completed. The demographics of each sample appear to match census and other know characteristics of Minnesota very well. As is characteristic of telephone surveys, women were slightly over-sampled: results are weighted to compensate for this.

The Minnesota State Lottery research department completed analysis of the data by using the SPSS statistical analysis computer package. Before analysis, each response was assigned a statistical weight based on the 2000 United States Census to compensate for the over sampling of women and certain age groups.⁷ Lottery Research Student Worker Allison Ogren and Lottery Research Analyst Brian Malecha compiled information in this report under the supervision of Don Feeney, Research and Planning Director for the Lottery. Feeney holds a Master of Science degree in statistics and is a member of the American Association of Public Opinion Research, the American Statistical Association and its section on Survey Research methods.

⁶ For further information of the Hagen-Collier method and other methods of respondent selection, see Lavrakas, Paul J. Telephone Survey Methods: Sampling, Selection, and Supervision, Sage Publications, Newbury Park, CA, 1987.

⁷ For a detailed discussion of stratified random sampling and weighting see Cochran, William G., Sampling Techniques, John Wiley & Sons, New York, 1977.