Transcript Slide 1

Exploring Demographic and Employment Characteristics of Employees
with Self-reported Gambling Problems
Margaret K. Glenn, EdD, CRC ;
Carolyn E. Hawley, Ph.D., CRC
West Virginia University
I: Using Exploratory Data Analyses in Addition to Conventional Null-Hypothesis-Testing (NHT)
Approaches to Explore Variables Influencing Employment Status
CHAID* Analysis helps investigators “re-think” how ordinal response ranges
are constructed. Here we see that when predicting the employment status of
people seeking help for problem gambling, “Age” is the most significant factor.
Employment Status
Node 0
Category
%
n
Employed 66
713
Unemployed 33
359
Total
(100 ) 1072
For people ages < 36, and 36- 55, “Co-morbidity” is the next most significant
factor with individual’s reporting pre-existing SA problems more likely to be
employed than those reporting MH conditions. However, for individuals with a
pre-existing Mental Health condition < 36, “Life Event” (defined as an event
that precipitated gambling and is believed by the individual to have triggered
the behavior), was the next most significant factor, while for ages 36-55,
“Family History” is significant . Individuals in this category with a family history
of PG or SA are less likely to be employed. Finally, for individuals 55-64,
women are more likely to be employed than men.
Current Age
Adj. P-value=0.0000, Chi-square=48.4229, df=2
18-35
Node 1
Category
%
Employed 74
Unemployed 26
Total
(56 )
36-55
Node 2
Category
%
Employed 62
Unemployed 38
Total
(30 )
n
447
158
605
Co-Morbidity
Adj. P-value=0.0410, Chi-square=6.0814, df=1
Mental Health Condition
Node 4
Category
%
Employed 77
Unemployed 23
Total
(35 )
Node 5
Category
%
Employed 68
Unemployed 32
Total
(21 )
Substance Abuse,
Node 6
Category
%
Employed 72
Unemployed 28
Total
(15 )
n
157
73
230
Triggering Life Events
Adj. P-value=0.0279, Chi-square=11.0229, df=1
Employment; Other; Death; Relationship
Node 10
Category
%
Employed 78
Unemployed 22
Total
(12 )
n
97
28
125
Node 3
Category
%
Employed 44
Unemployed 56
Total
(13 )
n
203
122
325
Co-Morbidity
Adj. P-value=0.0015, Chi-square=12.1081, df=1
Substance Abuse
n
290
85
375
56-64
n
117
46
163
n
86
76
162
Female
Node 8
Category
%
Employed 54
Unemployed 46
Total
(7 )
Male
n
41
35
76
Node 9
Category
%
Employed 33
Unemployed 67
Total
(6 )
n
22
44
66
Family History
Adj. P-value=0.0027, Chi-square=11.0282, df=1
Health
Node 11
Category
%
Employed 57
Unemployed 43
Total
(10)
Gender
Adj. P-value=0.0137, Chi-square=6.0814, df=1
Mental Health Condition
Node 7
Category
%
Employed 53
Unemployed 47
Total
(15 )
n
63
79
142
Family HX of Addiction; Family HX of Gambling
n
60
45
105
Node 12
Category
%
Employed 86
Unemployed 14
Total
(6 )
n
56
9
65
<missing>
Node 13
Category
%
Employed 62
Unemployed 38
Total
(9 )
n
61
37
98
Chi-squared Automatic Interaction (CHAID) Technique analysis:
•CHAID is a non-parametric analysis based on statistically recursive
partitioning algorithms. The CHAID Technique determines the relative
importance of each of the independent (predictor) variables in explaining
group membership in a categorical dependent (outcome) variable.
•Dendograms (i.e., classification trees) are utilized to display the relative
importance of statistically significant independent variables on the dependent
variable. The hierarchical nature of the CHAID dendograms provide a visual
depiction of variable interactions that may not be otherwise observable or
detected in traditional analytic procedures.
* The intention of this analysis is to determine factors that place people at risk
for developing gambling problems in order to develop targeted workplace
prevention and intervention programs.
II: Occupational Categories of Individuals
Seeking Help for Problem Gambling (PG)
Standard
Occupation
(SOC)
Category
SOC
#
Management
Business,
Financial
1113
N
%
58
9.3
Occupation/
Employment
Setting
Management
36
Finance/Insurance
17
Executive
Professional
15-29
141
22.5
Sales and
Office
31-39
41-43
103
137
16.4
21.9
66
Education
29
Computer/Informatio
n Technology
19
Healthcare
15
Professional/Scientif
ic/Technology
10
45-53
149
23.8
78
Gaming
13
Law Enforcement
8
Administrative
Waste Management
4
Sales
47
Clerical
35
Retail Clerk
28
Postal Employee
13
Wholesale/Retail
Trade
10
55-60
39
6.2
33
Manufacturing
35
Construction
29
Laborer
20
Industrial
23
March 29, 2008 * 2008 Annual American Counseling Association Conference, Hawaii
685
100.0
9
State Government
Employee
27
Federal
Government
Employee
10
Military
Total
4
Transportation/Ware
housing
Mining
Military
Government
2
Food/Hotel
Real Estate, Rental
and Leasing
Natural
Resources,
Construction
Production,
Transportati
on
5
Social Service
Arts/Entertainment
Service
N
2
68
5
CHAID* Analysis helps investigators “re-think” how
ordinal response ranges are constructed. Here we see
that when predicting the employment status of people
seeking help for problem gambling, “Age” is the most
significant factor. For people < 36, and 36- 55 “Comorbidity” is the next most significant factor with
individual’s reporting pre-existing SA problems more
likely to be employed than those reporting MH
conditions. However, for individuals with a pre-existing
Mental Health condition < 36, “Life Event” (defined as
an event that precipitated gambling and is believed by
the individual to have triggered the behavior), was the
next most significant factor; while for ages 36-55,
“Family History” is significant. Individuals with a family
history of PG or SA are less likely to be employed.
Standard
Occupation
(SOC)
Category
SOC
#
N
%
Managemen
tBusiness,
Financial
1113
58
9.3
Professional
Service
Sales and
Office
Natural
Resources,
Constructio
nProduction
,
15-29
31-39
41-43
45-53
141
103
137
149
22.5
16.4
21.9
23.8
Occupation/
Employment
Setting
N
Management
36
Finance/Insurance
17
Executive
5
Social Service
66
Education
29
Computer/Informati
on Technology
19
Healthcare
15
Professional/Scienti
fic/Technology
10
Arts/Entertainment
2
Food/Hotel
78
Gaming
13
Law Enforcement
8
Administrative
Waste
Management
4
Sales
47
Clerical
35
Retail Clerk
28
Postal Employee
13
Wholesale/Retail
Trade
10
Real Estate, Rental
and Leasing
4
Transportation/War
ehousing
33
Manufacturing
35
Standard
Occupation
(SOC)
Category
SOC
#
Management
Business,
Financial
1113
N
%
58
9.3
Occupation/
Employment
Setting
Management
36
Finance/Insurance
17
Executive
Professional
15-29
141
22.5
Sales and
Office
31-39
41-43
103
137
16.4
21.9
66
Education
29
Computer/Informatio
n Technology
19
Healthcare
15
Professional/Scientif
ic/Technology
10
45-53
149
23.8
78
Gaming
13
Law Enforcement
8
Administrative
Waste Management
4
Sales
47
Clerical
35
Retail Clerk
28
Postal Employee
13
Wholesale/Retail
Trade
10
55-60
39
6.2
33
Manufacturing
35
Construction
29
Laborer
20
Industrial
23
685
100.0
9
State Government
Employee
27
Federal
Government
Employee
10
Military
Total
4
Transportation/Ware
housing
Mining
Military
Government
2
Food/Hotel
Real Estate, Rental
and Leasing
Natural
Resources,
Construction
Production,
Transportati
on
5
Social Service
Arts/Entertainment
Service
N
2
68
5
Employment Status
Node 0
Category
%
Employed
66
Unemployed 33
Total
(100 )
n
713
359
1072
Current Age
Adj. P-value=0.0000, Chi-square=48.4229, df=2
18-35
Node 1
Category
%
Employed
74
Unemployed 26
Total
(56 )
36-55
Node 2
Category
%
Employed
62
Unemployed 38
Total
(30 )
n
447
158
605
Co-Morbidity
Adj. P-value=0.0410, Chi-square=6.0814, df=1
Mental Health Condition
Node 4
Category
%
Employed
77
Unemployed 23
Total
(35 )
Node 5
Category
%
Employed
68
Unemployed 32
Total
(21 )
Substance Abuse,
Node 6
Category
%
Employed
72
Unemployed 28
Total
(15 )
n
157
73
230
Triggering Life Events
Adj. P-value=0.0279, Chi-square=11.0229, df=1
Employment; Other; Death; Relationship
Node 10
Category
%
Employed
78
Unemployed 22
Total
(12 )
n
97
28
125
Node 3
Category
%
Employed
44
Unemployed 56
Total
(13 )
n
203
122
325
Co-Morbidity
Adj. P-value=0.0015, Chi-square=12.1081, df=1
Substance Abuse
n
290
85
375
56-64
Node 11
Category
%
Employed
57
Unemployed 43
Total
(10)
n
117
46
163
Family History
Adj. P-value=0.0027, Chi-square=11.0282, df=1
Health
Family HX of Addiction; Family HX of Gambling
n
60
45
105
Node 12
Category
%
Employed
86
Unemployed 14
Total
(6 )
n
56
9
65
Gender
Adj. P-value=0.0137, Chi-square=6.0814, df=1
Mental Health Condition
Node 7
Category
%
Employed
53
Unemployed 47
Total
(15 )
<missing>
Node 13
Category
%
Employed
62
Unemployed 38
Total
(9 )
n
61
37
98
n
63
79
142
n
86
76
162
Female
Node 8
Category
%
Employed
54
Unemployed 46
Total
(7 )
Male
n
41
35
76
Node 9
Category
%
Employed
33
Unemployed 67
Total
(6 )
n
22
44
66