David Casey*, Annik Mossière, Rob Williams, Nady el- Guebaly, David Hodgins, Garry Smith, Rob Williams, Don Schopflocher, & Rob Wood *Research Coordinator, Leisure, Lifestyle, Lifecycle Project (LLLP), Psychology Department, University of Calgary
Jan 04, 2016
David Casey*, Annik Mossière, Rob Williams, Nady el-Guebaly, David Hodgins, Garry Smith, Rob Williams, Don
Schopflocher, & Rob Wood
*Research Coordinator, Leisure, Lifestyle, Lifecycle Project (LLLP), Psychology Department, University of Calgary
Background on the Leisure, Lifestyle, Lifecycle Project (LLLP) Explain the biopsychosocial model
Describe: Adolescent Sample Measures
Present the results of logistic regression analysis for adolescents
Discuss the conclusions
Plans for the future: Examining patterns of relationship over three more collection
points Changes in gambling behavior over time
Cohort longitudinal study of gambling behavior Over 5 years, with 4 data collections Initial sample
Most recruited through Random Digit Dialing (RDD) Stratified by region of the province (urban & rural) 5 age groups (13-15, 18-20, 23-25, 43-45, 63-65) Divided into at-risk gamblers & general population
Data collection at Wave 1: Telephone, computer-based, & face-to-face interviews
Data collection at Wave 2: Web-based survey
Data collection at Wave 3: Just wrapped-up this month using web-based survey
Data collection at Wave 4: (in 12-16 months) Testing a biopsychosocial model of gambling
FAMILY HISTORY- Social & problem gambling- Substance use disorders- Psychiatric disorders- Deviance
COGNITIVE - Intelligence - Attentional Ability- Gambling fallacies- Coping Skills
FAMILY ENVIRONMENT- Parental behavior- Marital Status/conflict- Abuse experiences
EXTRA FAMILIAL ENVIRONMENT- Social Support- Friendships/peers- Religion/Spirituality- Ethnicity/Culture- Social organization
TEMPERAMENT/PERSONALITY- Impulsivity- Trait anxiety- Moral disengagement- Self-esteem
GAMBLING INVOLVEMENT- Frequency & Duration- Type & Range- Context
DEMOGRAPHICS- Religion- Age- SES- Family background- Ethnicity
EXTERNALIZING PROBLEMS- Alcohol use- Substance use- Tobacco use- Delinquent activity- Sexual activity
INTERNALIZING PROBLEMS- Depression- Anxiety
PREVENTION & TREATMENTBROADER SOCIO-CULTURAL FACTORS
- Availability of gambling; public attitudes; prevention programs; legislative changes; gambling knowledge
GAMBLING DISORDERS- Frequency & Duration- Type & Range- Context
BIOLOGICAL RISK- Neuropsychological functioning
- Frontal lobe- Neurotransmitter - DA (blood & saliva DNA)
- MAOI activity- Gender
STRESSORS- Physical health/disability- School/work- Familial/peer- Legal
Total Population Completes (N=1808)
N %
Age 13-15 Year Olds18-20 23-25 43-45 63-65
436315341402314
24.117.418.922.217.4
GenderMaleFemale
837971
46.353.7
LocationCalgaryEdmontonGrande PrairieLethbridge
754536224294
41.729.612.416.3
Marital Status (Adults Only)Single, Never Married Married/Common-lawDivorced /Separated/Widowed
570643156
41.647.0
11.4
Level of EducationLess than High SchoolCompleted High SchoolSome Technical/CollegeCompleted Tech/CollegeSome UniversityUniversity Degree
549279203225236315
30.415.411.212.513.117.5
Current Employment StatusNot Currently EmployedEmployed Part-TimeEmployed Full-Time
746430631
41.323.834.9
This talk will present findings from Wave 1 data only
Focus on adolescent sample
Examining relationship between gambling, family environment, religiosity, externalizing and internalizing problems
Non-Gambler Population (N = 196)
Gambler Population (N = 240)
Total Population(N =436)
n % n % n %
Age: 13 yrs 77 39.3 84 35.0 161 36.9
14yrs 71 36.2 76 31.7 147 33.7
15-16yrs 48 24.5 80 33.3 128 29.4
Gender: Male 91 46.4 144 60.0 235 53.9
Female 105 53.6 96 40.0 201 46.1
Location: Calgary 75 38.3 102 42.5 177 40.6
Edmonton 56 28.6 75 31.3 131 30.0
Grande Prairie
24 12.2 30 12.5 54 12.4
Lethbridge 41 20.9 33 13.8 74 17.0
Employment: Not Employed
158 80.6 176 73.3 334 76.6
Part OR Full-Time
38 19.4 64 26.7 102 23.4
Household Income: $0 TO $29,999
13 6.5 7 2.9 20 4.5
$30,000 TO $49,999
17 8.7 22 9.2 39 9.0
$50,000 TO $79,999
59 30.1 49 20.4 108 24.8
$80,000 OR Greater
107 54.6 162 67.5 269 61.7
Constructs from Figure 1 Construct Measure
Biological Risk Demographics Gender
Internalizing and Externalizing Problems
Psychopathology/Delinquent Activity/
Temperament/Personality
Child Behavior Checklist (CBC)
Alcohol, Substance & Tobacco Use
Canadian Community Health Survey (CCHS)
Cognitive Intelligence Wechsler Abbreviated Scale of Intelligence (WASI)
Family Environment Abuse Experiences Childhood Trauma Questionnaire (CTQ)
General Functioning/Family Support
Family Environment Scale
Extra-Familial Environment Social Support Lubben Social Network Scale (LSNS)
Religiosity Rohrbaugh Jessor Religiosity Scale (RJRS)
Culture York Ethnicity Scale
Social Organization Buckner Neighborhood Cohesion Scale (2 questions )
Stressors/Life Events Life Events Life Events Questionnaire
Physical Health SF-10 Health Survey
Gambling Involvement Frequency, Expenditure, Type, Range, Context, Motivation, & Knowledge
Canadian Problem Gambling Index (CPGI)
Attitude Gambling Attitude Questionnaire
Gambling Disorders Problem Gambling Fisher DSM-IV-MR-J
n Mean $ Spent
Lottery & Raffle Tickets 120 6.25
Instant Win Tickets 25 4.64
VLTs & Casinos 6 6.06
Private Games 94 10.84
Sport Betting 41 7.71
Other [Bingo, Horse Racing, High Risk Stocks]
18 62.22
n %
Never Gambled or Not in Last 12 Months
237 54.4
Gambled in Last 12 Months 199 45.6
Non-Gambler or Non Problem Gambler 333 76.4
Low Risk Gambler 72 16.6
Moderate Risk / Problem Gambler 31 7.0
First Step in analysis of adolescent gambling:
Univariate analysis were used to compare: Non-gamblers vs. Gamblers
Those significant at the univariate level were used in logistic regression analysis
Why Logistic Regression? Allows categorization into groups based on predictors Can use continuous and categorical variables
Data was weighted based on gender, age, and demographic location for Alberta
Bootstrap weights were used in the present analysis Refine the confidence interval in logistic regression
Male Adolescent Correlations
Female Adolescent Correlations
Total Correlations
Drug Use .26** .11 .17**
Alcohol Consumption .26** .18** .22**
CBC: Somatic .16* .14* .12**
Thought .10 .20** .15*
Attention .02 .19** .11*
Rule Breaking .24** .15* .21**
Aggressive .14* .07 .11*
Contact Friends
.18** .00 .10*
Anxious .09 .14* .08
Religiosity -.03 -.05 -.05
FES: Conflict .29** .10 .19**
Active/Recreational
.23** .08 .14**
Moral Religious
-.14* -.14* -.15**
Age .28** .02 .15**
Gender -.11*
Location .16* .18** .16**
Household Income .15* .07 .12**
Employment .06 .08 .08
Peer Gambling .32** .30** .31**
Sibling Gambling -.09 .09 .01
WASI IQ Score .00 .17** .09*
** . Correlation is significant at the 0.01 level (2-tailed) *. Correlation is significant at the 0.05 level (2-tailed)
Male Adolescents
Female Adolescents
Adjusted Odds Ratio (OR)(95% Confidence Intervals)
Non-Gambler (N = 95)
Gambler (N = 108)
Non-Gambler (N = 143)
Gambler (N = 91)
Male OR
p-value
Female OR
p-value
Religiosity 11.99 11.34 13.68 13.05 1.08(0.98,1.18)
.055 1.09 (0.97,1.21)
.013
FES - Conflict 2.11 3.47 2.51 3.11 1.27 (1.01,1.61)
.003
Active 5.75 6.71 6.33 6.64 1.52 (1.15,2.02)
.000 1.22(0.90,1.57)
.043
Moral 4.78 3.96 4.86 4.24 .78 (0.61,1.00)
.020 .78 (0.57,1.04)
.005
Peer Gambling 2.24 9.93 1.19 4.18 1.05 (1.01,1.10)
.010
Age 13.74 14.21 13.89 13.97 1.53 (0.89,2.62)
.051
Drug Use .02% .15% .09% .16& .18 (0.03,1.19)
.044
Male Adolescents
Female Adolescents
Adjusted Odds Ratio (OR)(95% Confidence Intervals)
Non-Gambler (N = 95)
Gambler (N = 108)
Non-Gambler (N = 143)
Gambler (N = 91)
Male OR
p-value
Female OR
p-value
CBC –Attention 4.20 4.32 3.82 4.79 1.26 (0.94,1.64)
.012
Thought
3.54 4.27 3.61 5.49 1.22 (0.98,1.41)
.006
Rule Break
3.09 4.53 2.81 3.95 1.20 (0.93,1.49)
.036
Aggressive
6.22 7.98 6.28 7.79 0.87 (0.75,1.03)
.040
Religiosity 11.99 11.34 13.68 13.05 1.08 (0.98,1.18)
.055 1.09 (0.97,1.21)
.013
FES – Active 5.75 6.71 6.33 6.64 1.52 (1.15,2.02)
.000 1.22 (0.90,1.57)
.043
Moral 4.78 3.96 4.86 4.24 .78 (0.61,1.00) .020 .78 (0.57,1.04) .005
Location 89.9% 95.9% 93.2% 99.7% 0.04 (0.01, 0.15)
.088
Household Income *
100341 115099 93035 118341 1.00 (1.00,1.00)
.029
WASI IQ Score 107.70 107.74 104.48 107.82 1.04 (0.99,1.08)
.014
* Household Income in Thousands, rounded to the nearest dollar
Compared to non-gamblers, male gamblers were: Older Identified more conflict in their family Involved in more activity and recreation with their family More likely to have used drugs in the past 12 months More likely to have peers who also gambled
Compared to non-gamblers, female gamblers: Scored higher on attention problems, thought problems, rule-
breaking, and aggression Were more involved with activity and recreation with their
family Came from households with a higher annual income Scored higher on the measure of intelligence
Moral and religious beliefs were protective factors for both adolescent males and females
Both males and females – less likely to gamble if they identified having strong moral and religious beliefs, either themselves or their families
Adolescents identified as having strong moral and religious beliefs associate gambling with immoral behavior, and thus it would be seen negatively, by their families and communities, for them to partake in the activity
Compare findings with data collected at Waves 2, 3, & 4:
Do the results remain consistent or change?
Are there still gender differences?
Availability to consider other constructs:
Do they help distinguish between non-gamblers and gamblers?
Examining change in gambling behavior over time:
How does the pattern change over 5 years? LLLP Waves 2-4: provide opportunity to examine changes in
behaviors associated with : Gambling Changes as they mature into young adulthood Changes in family environment Changes in moral and religious beliefs Other lifestyle altercations
What changes occur once they are of legal age to gamble? Important to examine changes in intensity of gambling over the
years, and expenditure in relation to their psychological health
The influence of other risky behaviors, such as the use of drugs and alcohol, will be important to consider as these adolescents mature into adulthood
Findings highlight interesting factors related to gambling behavior among a sample of adolescent males and females
Identifying the relationship between adolescent gambling, their peers gambling behavior, family, religion, and alcohol and substance abuse
can offer insight into guiding treatment approaches adolescents with gambling problems
Agencies could use these findings to: educate the public about the dangers of gambling creating awareness of the potential harm it can have on youth the role that religiosity, family, peers, and substance use can play
Legislators could develop more effective laws and policies regarding age restrictions associated with gambling, advertisement regulations, and access to gambling
David Casey, [email protected]
University of CalgaryPsychology Department
We Would Like to Acknowledge Funding for this Study from the Alberta Gambling Research
Institute (AGRI)
Gambling in Alberta 82% of adults gambled in previous year
Few studies of determinants of gambling & disordered gambling
Interested in better understanding: Factors that promote responsible gambling Factors that make some susceptible to problem gambling
Low prevalence of problem gambling requires over-sampling of at-risk groups
Longitudinal study as optimal methodology Over 5 years, with 4 data collections
Recruited through Random Digit Dialing (RDD) at 4 locations: Calgary Edmonton Grande Prairie (and surrounding communities) Lethbridge (and surrounding communities)
Start and end for data collection was staggered between sites Start: Feb 8, 2006 to Mar 20, 2006 End: Aug 26, 2006 to Oct 21, 2006
Recruited the following: Participants from the general population Participants at-risk of developing gambling problems
Based on frequency & amount of gambling
For all participants who met the criteria for age, residence, etc., there was the following at Wave 1:
Telephone interview by subcontract Adult interviews (~ 45 minutes) Adolescent interviews (~ 30 minutes) Majority of demographic & gambling questions
Face-to-face interview by RA’s Adult interviews (~ 3 hrs) Adolescent interviews (~ 2 hrs) Parent interviews (~ 40 minutes)
Response rate <10%
Differences for males and females
Pattern of relationship with predictor variables was different
Logistic regressions were separate for males and females
Constructs from Figure 1
Construct Measure
Internalizing and Externalizing
Problems
Individual Risk Taking Risk Taking
Family Environment Parental Monitoring Parental Monitoring (Adolescent & Parent)
Extra-Familial Environment
Social Support Loneliness and Social Dissatisfaction Scale (16 items)
Stressors/Life Events Coping Coping Inventory for Stressful Situations (CISS)
Physical Health – Eating Disorders
Eating Disorder Examination Questionnaire (EDE-Q 6.0)
Physical and Mental Health Wellness Index
Difficulty to recruit using Random Digit Dialing: Used Computer-Assisted Telephone Interview (CATI) Call display; Blocking; “Do not call” lists Saturation of the saturation
Difficulty to recruit at-risk or high-risk gamblers Supplemental recruitment techniques N=30 only!
Media release; Ads in local papers; Posters in casinos; “Snowball” e-mail
Telephone to face-to-face interview loss: Some did not feel $75 was enough incentive Booming economy vs. recession
Ability to look at changes in patterns of gambling behavior over time
3 more data collections:
Wave 2 completed from Nov. 2007 to Jun. 2008
Wave 3 started in Jul. 2009 to April 2010
Wave 4 will begin in the Winter of 2010
Wave 2 to 4 participants will complete web-based surveys
Gambling behavior will be tracked over all 5 years
Constructs associated with biological, psychological, & social factors will also be tracked