Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
46
Chalita Khirirat, M.D., Teerapat Teetharatkul, M.D., Jaturaporn Sangkool, M.D.
Department of Psychiatry, Faculty of Medicine, Prince of Songkla University, Hat Yai, Songkhla 90110, ailand.
Life Assets and Substance Use of High School
Students in Songkhla
ABSTRACT
Objective: To examine the association between life assets (including demographic factors) and substance use among
high school students.
Materials and Methods: In this cross-sectional study, 1,713 participating students were asked to ll out questionnaires.
Data were analyzed using descriptive statistics (e.g., mean, frequency and standard deviation); associating factors
were analyzed via univariate analysis and ordinal logistic regression.
Results: Among the participants, 67.5% were females with a mean age of 16.4 ± 0.96 years. On average, the participants
have a good level of overall life assets, except for community power. ose with a mean age of 14.3 years engaged in
substance use early, with peer inuence being a leading cause (8.1%). 48.0% of the participants had used substance
in their lifetime, and 74.7% of them had used only 1 substance, with alcohol being the most prevalent (97.7%). An
excellent level of participant life assets, especially wisdom power, was negatively associated with substance use (OOR
= 0.48). Other signicant factors that discouraged substance use were being an only child (OOR = 0.75), having a
high cumulative grade point average (GPA) (OOR = 0.63), and belonging to a two-parent family. (OOR = 0.79)
Conclusion: is study arms a negative relationship between an excellent level of life assets, especially wisdom
power, and substance use. Demographic factors like being an only child, having a high GPA, and belonging to
two-parent households also showed a strong negative association with substance use.
Keywords: Life assets; substance; student; high school (Siriraj Med J 2021; 73: 46-54)
Corresponding author: Jaturaporn Sangkool
E-mail: sjaturap@medicine.psu.ac.th
Received 14 September 2020 Revised 8 October 2020 Accepted 9 October 2020
ORCID ID: http://orcid.org/0000-0002-6395-1394
http://dx.doi.org/10.33192/Smj.2021.07
INTRODUCTION
Substance use is a substantial public health problem
worldwide. e United Nations Oce on Drugs and
Crime estimates that 271 million people around the
globe used substances in 2016. Moreover, the number
of substance users who suered from substance use
disorders and who died from substance use was 35
million and more than half a million, respectively.
1
Substance abuse is a social problem in ailand as well.
According to surveys, 2.9 million Thai people were
substance abusers in 2016
2
(youths aged 12-15 years in
Bangkok are exposed to substance use early)
3
, and 50% of
patients at rehabilitation centers were youths who tended
to use combinations of substances.
4
In addition to the
propensity for developing health problems, adolescents
who use substances excessively are highly susceptible to
having irritability and psychosis, leaving schools, having
broken families, and committing crimes.
4,5
erefore, it
is important to identify adolescents at risk for substance
use and provide an eective early intervention for them.
Khirirat et al.
Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
47
Original Article
SMJ
Prior studies have examined factors that are negatively
associated with adolescent substance use. ese protective
factors-or assets-include academic achievement
6
, family
support
6,7
, positive peers
8,9
, and adult role models.
7
In
particular, adolescents with high academic aspirations
6
,
supportive parents
7
, and low substance use among friends
9
are less inclined to engage in substance use.
Apart from these protective factors, life assets are
viewed as positive supports and strengths that may
prevent adolescents from engaging in risk behaviors, e.g.
substance use. In the United States, the Developmental
Assets Framework was developed to evaluate life assets
using 40 indicators.
10
Based on this framework, many
studies have reported that individuals with many life assets
were less likely to use cigarettes, alcohol, and marijuana
in the past 30 days.
11
In addition, life assets may promote
positive behaviors such as high academic achievement,
the use of safety belts, and engaging in aerobic exercise.
11
In ailand, using the same framework
10
, life assets have
been assessed using 48 indicators that are categorized
into 5 domains: self-power, family power, peer power,
wisdom power, and community power.
5,12
ai high school
students have been reported to pass the overall criteria
for life assets, except community power.
13–15
However,
there is limited evidence showing how life assets may
prevent substance use among students in ailand.
The objective of this study was to examine the
relationship between life assets as well as other demographic
factors and substance use in high school students in
Songkhla province, an endemic area of substance use
in Southern ailand.
16
MATERIALS AND METHODS
e Ethics Committee of the Faculty of Medicine,
Prince of Songkla University approved this cross-sectional
study (REC.62-032-3-4). It was conducted from June
to September, 2019. e sample size was calculated by
G*Power that had an eect size as 0.1, the Alpha as 0.05,
the Power as 0.8 and the Degree of Freedom (df) as 4.
Among 1,713 students studying in grades 10-12 from 7
high schools in Hat Yai, which were selected randomly
using cluster sampling and probability proportional to size.
To be enrolled in the survey, the students needed to be
able to both understand and complete the questionnaires,
while students on a leave of absence were excluded.
Methodology
e randomly selected students were rst introduced
to the rationale of the study, and then they were informed
about the acquired information as well as the protection
of their condential information. If the students agreed
to participate in the surveys, they would take a few
minutes to complete the questionnaires without the
requirement for their signatures in order to ensure their
anonymity. Aer their completion, the questionnaires
were placed in boxes in front of the classroom by the
students themselves. e results were returned to the
participants’ e-mail addresses. All participants had the
right to request a withdrawal at any time and without
repercussion.
Instruments
e questionnaires consisted of 4 parts:
1) Demographic data comprised gender, age, ethnicity,
academic year, cumulative grade point average (GPA),
child order, religion, area of residence, type of residence,
people living with, parent’s marital status, and parent’s
level of education.
2) e Life Assets Questionnaire (youth version)
consisted of 48 items enquiring about self-power (15
items), family power (8 items), wisdom power (11 items),
peer power (6 items), and community power (8 items).
For each item, the responses ranged from “0” (none) to
“3” (regularly). e total scores were categorized into 4
levels: excellent (≥80.0%), good (70.0-79.9%), moderate
(60.0-69.9%), and low (<60%), which was considered
to indicate failure. In this questionnaire, a Cronbach’s
alpha is 0.89.
12
3) Substance use information included age at initiation
of use, reasons for use, and types of substance used.
4) e ASSIST-Lite ai version
17
consists of 8
questions related to illicit substances; there are a few
questions enquiring in more detail about each substance.
In total, it contains 19 items rated with “Yes” (score
one) or “No” (score zero). e cut-o score of a likely
substance use disorder (SUD) is 2, while the cut-o score
of alcohol is 3. e ASSIST-Lite has the AUC [0.8-1.0],
sensitivity [0.8-1.0] and specicity [0.7-0.8].
18
In this
study, we use the ASSIST-Lite to assess the past-3-month
substance use among students.
Statistical analysis
e data were analyzed using descriptive statistics, e.g.
percentage, frequency, average, and standard deviation.
e factors associated with substance use were scrutinized
via univariate analysis. If the p-value from the univariate
analysis was less than 0.2, those factors were analyzed
using an ordinal logistic regression. ose factors with a
p-value of less than 0.05 were indicated as the statistically
signicant factors.
Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
48
RESULTS
Demographic data
Of the 1,713 participants, 67.5% were female with
mean age of 16.4 ± 0.96 years (range 15-19 years). e
students studied in grades 10 (31.9%), 11 (33.0%); and
12 (35.0%). e majority of them was Buddhist (89.6%),
had siblings (80.2%), had GPAs in the 3.00-4.00 range
(53.1%), lived in urban areas (71.8%), and lived in two-
parent households (65.0%).
Substances
Fig 1 shows the details of substance use categorized in
3 groups: students with a lifetime substance use, students
with a past-3-month occasional use, and students with a
likely substance use disorder (SUD) based on the past-
3-month use. e common substances in the lifetime
use category were alcohol (97.7%), smoking (21.6%),
and stimulants (8.8%). About three quarters (74.7%)
of participants used 1 substance, while the remaining
quarter (25.3%) were polysubstance abusers. e number
of male students was higher than that of females in terms
of using substances in combination.
Fig 2 shows that the proportions of participants
who reported substance use by grade were: grade 12
(38.4%), grade 11 (36.4%), and grade 10 (25.2%). 66%
of them were identied as female, 32.5% as male, and
1.5% as other.
As shown in Fig 3, approximately half (48.0%) of the
participants reported a lifetime substance use. 39.3% of
participants were past-3-month substance users and 8.7%
of participating students were past-3-month substance
non users. Fig 4 shows the past-3-month substance users.
63.4% of them reported occasional substance users, and
36.6% were students with likely SUD.
e average age of the participating students at the
initiation of substance use was 14.3 years. Fig 5 shows
that the reasons for engaging in substance use were peer
inuence (8.1%), sadness or stress (7.0%), and curiosity
(6.5%).
Fig 1. Substances used in the 3 groups: lifetime substance users, past-3-month occasional users, and past-3-month users with likely SUD
(n=818).
Khirirat et al.
Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
49
Original Article
SMJ
Fig 5. e reasons of substance use.
Fig 2. Academic year and substance use.
Fig 3. Lifetime and past-3-months substance users.
Fig 4. Past-3-months substance users.
Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
50
Life assets
On average, the participants had a good level of life
assets; 81.9% of them met the criteria for life assets. In
particular, they had an excellent level of family power,
a good level of self-power, a moderate level of peer and
wisdom power, and a low level of community power.
Table 1 presents the data related to the overall life
assets as well as per the 5 life asset domains according
to the 3 subgroups: non-users, past-3-month occasional
users, and past-3-month users with likely SUD. In terms
of the life assets, the majority of non-users and past-
3-month occasional users had the good level, while past-
3-month users with likely SUD had the moderate level.
e association between life assets, demographic data,
and substance use
Tables 1 and 2 show the univariate analysis results
of the demographic data and life assets according to the
3 student groups: non-users, past-3-month occasional users,
and past-3-month users with likely SUD. e variables
that were analyzed using an ordinal logistic regression
were gender, siblings, GPA, type of residence, people
living with, and life assets (self-power, family power,
and wisdom power). e majority of non-users, past-
3-month occasional users and past-3-month users with
likely SUD had similar trends in terms of demographic
factors; for example, the majority of the 3 subgroup lived
in two-parent households, had high GPA, and had an
excellent level of self-power.
As shown in Table 3, students with an excellent level
of life assets were less likely to engage in substance use,
and the related factors that were found to discourage
substance use statistically were being an only child, having
a GPA the range between 3.00 and 4.00, and belonging
to a two-parent family (P-value < 0.05). Moreover,
a negative relationship between having an excellent level
of wisdom power and substance use was detected (OOR
= 0.48, 95% CI 0.35-0.66, P-value < 0.001).
DISCUSSION
is research examines the association between life
assets and substance use in ai students. Our results
suggest a lower prevalence of substance use among
students with an excellent level of life assets (especially
wisdom power). According to previous studies in the
United States
8,11
, the number of assets was negatively
related to alcohol and substance use, and youths who
had any 1 of the assets were approximately 1.5 to 3
times more likely to report nonuse of substances. Our
ndings support that the demographic factors such as
being an only child, belonging to two-parent households,
and having a high GPA are statistically found to deter
students from engaging in substance use. e studies in
Kenya
19
, California
20
and ailand
21-24
also suggest that
students who are an only child, who belong to two-parent
households, and who have a high academic performance
may be less likely to use substances. Possible reasons may
be the protection and attention they receive from one or
both parents
19
, close parental monitoring
20
, their sound
understanding of the negative consequences of substance
use
24
, a good sense of competency, and a healthy level
of self-control.
23
According to our results, the students
met the overall criteria of life assets; the only exception
was community power. Consisting with past surveys in
ailand
13-15
, ai students also met the overall criteria of
life assets, except the community power. ese results may
be explained by the evolution of individualist societies.
Furthermore, 48% of the participating students
reported engaging in lifetime substance use, and 39.3%
were past-3-month substance users, of whom 63.4%
participated in occasional substance use, and the remaining
36.6% were participants with likely SUD. ese results
mirror the ndings of past studies, which have reported
that students tend to try substances rather than use them
regularly.
25
Alcohol was the most commonly used substance.
According to Bangkok survey
3
, ai teenagers easily
access to alcohol because two-thirds of them purchased
alcohol by themselves.
In addition, it was found that, on average, the
students initiated their use of substances at the age of 14.3
years due to primarily peer inuence, stress or sadness,
and curiosity. Consisting studies in Colorado
26
, peer
relationships are positively associated with adolescent
substance use. In ai surveys
3
, teenagers drank alcohol
for social purposes; drank alcohol with peers and drank
alcohol alone when they had life problems. Adolescents
who have peers that use substances and those experiencing
high stress levels tend to use substances because they
may observe and decide to imitate close their friends’
negative behaviors
21,27
or substances as self-medication
for mood-altering purposes.
26
In conclusion, both researchers and healthcare
practitioners may nd the results of our study useful for
the development in preventing substance use among high
school students. Finally, national public health policies
should focus on developing interventional strategies that
aim to boost the students’ life assets, and in particular,
to promote education related to the dangers of engaging
in substance use.
Khirirat et al.
Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
51
Original Article
SMJ
TABLE 1. Univariate analysis of life assets and their 5 domains among past-3-month substance users.
Life assets 0.01
Failure 310 (18.1) 146 (16.5) 94 (22.1) 48 (19.6)
Moderate level 482 (28.1) 241 (27.2) 121 (28.5) 75 (30.6)
Good level 525 (30.6) 263 (29.7) 134 (31.5) 72 (29.4)
Excellent level 396 (23.1) 236 (26.6) 76 (17.9) 50 (20.4)
Self-power 0.13
Failure 118 (6.9) 63 (7.1) 28 (6.6) 20 (8.2)
Moderate level 354 (20.7) 162 (18.3) 108 (25.4) 50 (20.4)
Good level 458 (26.7) 236 (26.6) 109 (25.6) 65 (26.5)
Excellent level 783 (45.7) 425 (48.0) 180 (42.4) 110 (44.9)
Family power 0.03
Failure 224 (13.1) 103 (11.7) 61 (14.4) 37 (15.1)
Moderate level 1163 (9.5) 72 (8.1) 46 (10.8) 29 (11.8)
Good level 317 (18.5) 152 (17.2) 86 (20.3) 49 (20.0)
Excellent level 1006 (58.7) 557 (63.0) 231 (54.5) 130 (53.1)
Wisdom power <0.001
Failure 541 (31.6) 251 (28.4) 166 (39.2) 88 (35.9)
Moderate level 502 (29.2) 253 (28.6) 123 (29.0) 70 (28.6)
Good level 291 (17.0) 141 (16.0) 63 (14.9) 51 (20.8)
Excellent level 377 (22.0) 239 (27.0) 72 (17.0) 36 (14.7)
Peer power 0.29
Failure 410 (23.9) 224 (25.3) 98 (23.1) 51 (20.8)
Moderate level 467 (27.3) 228 (25.7) 128 (30.1) 62 (25.3)
Good level 390 (22.8) 197 (22.2) 101 (23.8) 60 (24.5)
Excellent level 446 (26.0) 237 (26.7) 98 (23.1) 72 (29.4)
Community power 0.33
Failure 939 (54.8) 477 (54.0) 250 (58.8) 129 (52.7)
Moderate level 257 (15.0) 131 (14.8) 61 (14.4) 38 (15.5)
Good level 243 (14.2) 122 (13.8) 62 (14.6) 37 (15.1)
Excellent level 271 (15.8) 154 (17.4) 52 (12.2) 41 (16.7)
Number of users (%)
Life assets Overall Lifetime non Past-3-month Past-3-month P-value
domains Participants substance users occasional users users with likely
(n=1713) (n=886) (n=425) SUD (n=245)
Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
52
TABLE 2. Univariate analysis of demographic characteristics among past-3-month substance users.
Gender 0.002
Male 539 (31.5) 271 (30.6) 117 (27.5) 98 (40.0)
Female 1157 (67.5) 610 (68.8) 304 (71.5) 142 (58.0)
Other 17 (1.0) 5 (0.6) 4 (0.9) 5 (2.0)
Siblings 0.07
Only child 314 (18.3) 178 (20.3) 64 (15.2) 41 (17.1)
Siblings 1373 (80.2) 697 (79.7) 356 (84.8) 199 (82.9)
GPA <0.001
<3.00 450 (26.3) 200 (28.4) 120 (35.9) 83 (41.7)
3.00-4.00 910 (53.1) 505 (71.6) 214 (64.1) 116 (58.3)
Residence type 0.13
House 1497 (87.4) 749 (85.3) 338 (80.3) 199 (81.2)
Dormitory 153 (8.9) 68 (7.7) 48 (11.4) 28 (11.4)
Other (condominium) 51 (3.0) 61 (6.9) 35 (8.3) 18 (7.3)
People living with (could choose more than one answer)
Two-parent household 1114 (65.0) 602 (67.9) 270 (63.5) 139 (56.7) 0.004
One-parent household 335 (19.6) 168 (19.0) 79 (18.6) 58 (23.7) 0.21
Siblings 904 (52.8) 474 (53.6) 226 (53.3) 122 (49.8) 0.57
Friends 58 (3.4) 16 (1.8) 23 (5.4) 15 (6.1) <0.001
Alone 59 (3.4) 28 (3.2) 19 (4.5) 11 (4.5) 0.4
Parent’s marital status 0.43
Married 1263 (73.7) 668 (75.5) 310 (72.9) 171 (69.8)
Separated 102 (6.0) 53 (6.0) 26 (6.1) 14 (5.7)
Divorced 253 (14.8) 117 (13.2) 69 (16.2) 46 (18.8)
Widowed 93 (5.4) 47 (5.3) 20 (4.7) 14 (5.7)
Number of users (%)
Demographic data Overall Lifetime non Past-3-month Past-3-month P-value
Participants substance occasional users with
(n=1713) users users likely SUD
(n=886) (n=425) (n=245)
Khirirat et al.
Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
53
Original Article
SMJ
Limitations
ere are two main limitations in this study. First,
this cross-sectional survey cannot show a clear causal
relationship between the students’ life assets and substance
use because it employed only self-reporting questionnaires.
ere is a real possibility that the participants might
have not provided truthful information regarding their
substance use due to it being considered a socially
unacceptable behavior. Second, the study population
was randomly sampled only from students in Southern
ailand; therefore, these ndings might not be valid for
all ai students throughout the country. Furthermore,
the majority of the participants were female students,
who tend to be less likely to use substances.
Implications and future recommendations
Further studies should examine the life assets of
students in alternative schools as well as those pursuing
non-formal education. us, multi-center cohort studies
are recommended.
CONCLUSION
is study investigated the association between
individual life assets (including demographic factors) and
the use of substance. Its ndings revealed a statistically
positive relationship between life assets and substance
non-use. e students with an excellent level of life assets,
especially wisdom power, were signicantly less likely
to use substances. e related factors that discouraged
substance use were being an only child, having a high
GPA, and belonging to two-parent households.
ACKNOWLEDGMENTS
This study was fully funded by the Faculty of
Medicine, Prince of Songkla University, ailand. e
authors sincerely appreciate the invaluable role Mrs. Nisan
Werachattawan and Ms. Kruewan Jongborwanwiwat
played in the data analysis. e authors also would like
to express their gratitude to the participating students.
REFERENCES
1. United Nations Oce on Drugs and Crime. World drug report
2019. Vienna: United Nations publication, 2019.
2. Kanato M, Assanangkornchai S, Charoenratana S. National
household survey, estimation of population related with substance
abuse. Bangkok: Jaransanitvong Press, 2016.
3. Ruangkanchanasetr S, Plitponkarnpim A, Hetrakul P, Kongsakon
R. Youth risk behavior survey: Bangkok, ailand. J Adolesc
Health 2005;36:227-35.
4. Choosunthia W, Pinitsuntorn S. e impact of drug addiction
on patients admitted to anyarakhospital, Udonthani: a case
study. CHDKKU 2017;5:523-33.
5. Tripathi S, Musikaphan W. Quality of children’s life 2013. 1
st
ed. Bangkok: Appa printing group, 2013.
6. Kodjo CM, Klein JD. Prevention and risk of adolescent substance
abuse: the role of adolescents, families, and communities.
Pediatr Clin North Am 2002;49:257-68.
7. Beebe L, Vesely S, Oman R, Tolma E, Aspy C, Rodine S.
Protective assets for non-use of alcohol, tobacco and other
drugs among urban American Indian youth in Oklahoma.
Matern Child Health J 2008;12:82-90.
8. Oman R, Vesely S, Aspy C, McLeroy K, Rodine S, Marshall L.
e potential protective eect of youth assets on adolescent
alcohol and drug use. Am J Public Health 2004;94:1425-30.
9. Hussong A. Dierentiating peer contexts and risk for adolescent
substance use. J Youth Adolescence 2002;31:207-20.
TABLE 3. Multivariate analysis of demographic factors and life assets among past-3-month substance users.
Factors Ordinal odds ratio 95% CI P-value
Only child 0.75 0.55-1.00 0.03
GPA 3.00-4.00 0.63 0.50-0.79 <0.001
Two-parent household 0.79 0.63-0.99 0.02
Life assets
Moderate level 0.78 0.57-1.07 0.06
Good level 0.79 0.58-1.08 0.07
Excellent level 0.62 0.44-0.87 <0.001
Volume 73, No.1: 2021 Siriraj Medical Journal
https://he02.tci-thaijo.org/index.php/sirirajmedj/index
54
10. PL Benson, EC Roehlkepartain. Tapping the power of community:
building assets to strengthen substance abuse prevention.
Search Institute Insights & Evidence 2004;2:1-14.
11. Murphey D, Lamonda K, Carney J, Duncan P. Relationships
of a brief measure of youth assets to health-promoting and
risk behaviors. J Adolesc Health 2004;34:184-91.
12. Tripathi S, Sungthong P, Salachan S. User Guide on Life
Assets Inventory for ai Children and Youth (youth version).
2
nd
ed. Bangkok: Appa printing group, 2010.
13. Srichantr-intr N, Ketumarn P, Tripathi S. Life assets in
academically talented students: case study of the participants
in international mathematics and science Olympiad 2013. J
Psychiatr Assoc ailand 2014;59:151-62.
14. Tripathi S. e national survey of life assets (positive youth
development) among secondary school students in ailand.
J Adolesc Health 2013;52:s6.
15. Petchruschatachart U, Balthip K, Piriyakoontorn S. Life’s assets
and quality of life of Thai junior high school adolescent,
Songkhla province. Songklanagarind J Nurs 2016;36:55-69.
16. Oce of the Narcotics Control Board: Ministry of Justice.
Report of Illicit drug situation in 7 provinces of southern
ailand. Bangkok: Ministry of Justice, 2017.
17. Assanangkornchai S, Tantirungsi N. ASSIST, ASSIST-Lite,
ASSIST-Y and e-ASSIST. 1st ed. Songkhla: Prince of Songkla
University, 2014.
18. Ali R, Meena S, Eastwood B, Richards I, Marsden J. Ultra-rapid
screening for substance-use disorders: e Alcohol, Smoking
and Substance Involvement Screening Test (ASSIST-Lite).
Drug and Alcohol Dependence 2013;132:352-61.
19. Mukangi A. e role of birth order in substance related disorders.
AJOL 2010;2:221-37.
20. Hemovich V, Lac A, Crano WD. Understanding early-onset
drug and alcohol outcomes among youth: e role of family
structure, social factors, and interpersonal perceptions of use.
Psychol Health Med 2011;16:249-67.
21. Hawkins J, Catalano R, Miller J. Risk and protective factors for
alcohol and other drug problems in adolescence and early
adulthood: implications for substance abuse prevention. Psychol
Bull 1992;112:64-105.
22. Chinakate S. Factors affection to preventive behavior on
drug among the vocational certicate students in college under
the provincial vocational education of Ratchaburi [esis].
Nakhon Pathom: Silpakorn University; 2010.
23. Schulenberg J, Bachman JG, O’Malley PM, Johnston LD. High
school educational success and subsequent substance use: a
panel analysis following adolescents into young adulthood. J
Health Soc Behav 1994;35:45-62.
24. Chagphimai C, Sritanasal P. Self-defense behaviors from drugs
of students at King Mongkut’s University of technology north
Bangkok [esis]. Bangkok: King Mongkut’s University of
Technology North Bangkok; 2001.
25. Srisatchang A. Knowledge, understanding, and factors related
to students’ drug abuse at Prince of Songkla University [esis].
Songkhla: Prince of Songkla University; 2002.
26. Whitesell M, Bachand A, Peel J, Brown M. Familial, social,
and individual factors contributing to risk for adolescent
substance use. J Addict 2013;2013:1-9.
27. Bandura A. Inuence of models’ reinforcement contingencies on
the acquisition of imitative responses.J Pers Soc Psychol
1965;1:589-95.
Khirirat et al.
