Luận văn thạc sĩ Quản lý công: Adolescent violence and some related factors, findings from the study in Hai Duong province, Vietnam

Nguyễn Mạnh Hùng Time of thesis presentation: 12-13 December 2011 Objective: Addressing the situation of violence and identifying risk factors relating to adolescent violent behaviors in

UPPSALA UNIVERSITET &

UNIVERSITY OF ECONGMICS & BUSINESS, VNUH >)

ea a te co

VNU-UEB UPPSALA Tum <<

UNIVERSITET

MASTER THESIS OF MPPM

ADOLESCENT VIOLENCE AND

SOME RELATED FACTORS,

FINDINGS FROM THE STUDY

IN HAI DUONG PROVINCE, VIETNAM

Authors (in alphabet order): Tran Thu Huong

Nguyễn Thị Oanh

Nguyễn Tuấn Thanh

Supervisor: Dr Lars-Torsten Eriksson Local Supervisor: Dr Nguyễn Mạnh Hùng

Class: MPPM INTAKE 3 - Group 1

Hanoi, January - 2012

SS EEE =

We would like to express sincere gratitude towards our supervisors: Dr.

Nguyễn Mạnh Hùng and Dr Lars-Torsten Eriksson as well as all teaching staff who have been the inspiration throughout our learning as well as in writing this thesis, for the never-ending hours spent on guiding us on the right path, the great support and the valuable feedback along the way.

We also would like to thank staff members working at the Master of Public Management Program of the Economics University, Hanoi National University, affiliated with the Uppsala University, Sweden for their full support during the course and the progress of completing this thesis.

We would like to thank the Hai Duong Women’s Union, volunteers and any other persons involved in helping us in collecting data of the thesis.

Our team would like to express deep thanks to class leaders, especially

to the deputy class leader, who is always there by our side to encourage and

support the team’s work This paper would not be accomplished without her

Group I - MPPM -Intake 03

January, 2012

Title of the thesis:

Adolescent violence and some related factors, findings from the study in Hai Dương province, Việt Nam

Level: Final assignment for Master Program in Public Management.

Authors: Tran Thu Huong, Nguyễn Thị Oanh, Nguyễn Tuấn Thanh

Instructors:

Dr Lars-Torsten Eriksson and Dr Nguyễn Mạnh Hùng

Time of thesis presentation: 12-13 December 2011

Objective:

Addressing the situation of violence and identifying risk factors relating

to adolescent violent behaviors in Phú Thứ Commune and Minh Tân

Town, Kinh Môn District, Hải Duong Province.

Methodology:

The cross-sectional descriptive study is performed for the sample of 392 literate adolescents from 13 to 18 years old, living with both father and mother The study applies: the self-reported questionnaires for data collection; and the social-ecological model for analyzing the relationship between potential risk factors at individual/peer/family domains to

identify risk factors related to violent behaviors in adolescents Data is

analyzed using bivariate and multivariate statistical techniques (crosstabs

and logistic regression analysis).

Of the sample, 33.3% of male adolescents have injured others, higherthan female adolescents with just 21.0% And 64.5% of adolescentsliving in urban areas have injured others, higher than in rural areas with14.0% The relationship between negative and improper behaviors of

friends/family and adolescent violent behaviors is not clear.

Characteristics of adolescents most likely to injure others included:

smoking, drinking, living in urban areas, in wealthier families and have higher rate of being hurt by others.

Adolescent violence is relatively prevalent in Vietnam In term of

policy-making, although Vietnam has developed a rather comprehensive legal framework, it only addresses adolescent fundamental rights Adolescent violence is among emerging issues that require more attention The related risk factors show that national policies for prevention programs should include strategies to reduce violence in this population group.

Limitations and proposed next steps

Because of applying the descriptive, cross-sectional method, deploying the research in a narrow scope (Phú Thứ commune and Minh Tân town), selecting research subjects randomly to match with the study’s contexts,

therefore its results will not cover every details in adolescent violence situation in Vietnam 22 risk factors have been studied here only reflect a fundamental characteristics about local adolescents in association with

influences from their living environment Many other risk and protective factors have not yet examined its’ relationship with the adolescent

violence situation.

A next research would apply the longitudinal study to fnd out how risk factors, which have been found in this study, associate with the adolescent violence situation in Hai Duong altering over the time, under the impact of social-economic development Its findings will finalise the construction of a suitable intervention model for adolescent violence at

and programs needs to be improved.

words:

Cross-sectional research, social-ecological model, risk factors, violence,

adolescents.

CHAPTER 1: LITERATURE REVIEW

1.1 Background definitions giờ 1.1.1, The concept of “adolescent” wd 1.1.2 The concept of “violence” a3) 1.2 Overview of the Vietnamese Government policies relating to protection and prevention of adolescent violence 4 1.2.1 Adolescent care, protection, education and violent situation in Vietnam 4 1.2.2 Contemporary limitations in Vietnamese policies towards adolescents 6 1.3 Modeling causes of adolescent violent behaviors cccccccereey 7

1.3.1 Causal analysis model

1.3.2 Systematize problems by the Social-Ecological model

1.4 Some researches on adolescent violence worldwide «1.0.0: 13

1.5, Objectives, research questions

1.6 Research objectives and questions.

3.2 Description of violent situation in the target group

3.2.1 Overview of violent situation

3.2.2 Individual domain - Individual behaviors related to adolescent violent

Situation

3.2.3 Peer domain - Close friend behaviors related to adolescent violent

SiiURHGT :z.etrerrrimearrrsnsmmi

3.2.4 Family - Family effects related to adolescent violent situation Da,

3.3 Understanding the association between adolescent violent situation and risk

factors (logistic regression analy§iS) :-: :-+-++>

CHAPTER 4: DISCUSSION ccsccsssssscsossersenessssnssesssssersersensensenccesenssensencensesoanonsennss 35

4.1 The research sample and research subjects

4.2 Violence situation of local adolescents

4.3 Refer to the implementation of present policies on caring, protecting and

promoting adolescent health development in Vietnam

CHAPTER 5: CONCLUSIONS & RECOMMEDATIONS

5.1, Violent situation of adolescents in research areas

5.2 Recommendations

REFERENCES

APPENDIX

Annex 1: Legislative documents in Vietnam for adolescents and youth 46

Annex 2: The research questionnaires

Figure 1.3.1: Causal analysis framework

Figure 1.3.2: Social-Ecological Model

Figure 1.5.1: Theoretical Framework for adolescent violence - 15

LIST OF TABLES

Table 3.1.1: Distribution of target group by gender, age group, resident

location and education.

Table 3.2.1.1: Distribution of adolescent violent situation by gender, age

group, resident location and education

Table 3.2.2.1: Distribution of adolescent violent situation by individual’s risk

TấDIE3.32.:0Modelsurmman hee eee số 31

Table 3:3:3:: Hosmer and,Lemeshow.Lest eat 6 3] Table 35314: Classification table n6 la si tối Sil)

Table 3.3.5: Logistic Regression Model of risk factors related to the adolescent violent situation

Over the world, caring for and improving physical and mental health of

adolescents is given highest priorities by the entire society [26, 27] In

Vietnam, up to date, many legal documents have been issued by the Government, the Prime Minister and Ministries in order to bring about a safe, highly protective and developing environment for adolescents and youth [1-4,

10, 13] As a result, youth development has been given great attention by

central and local agencies and people in general.

Apart from the formation of a legal framework that is enabling care, protection and improvement of adolescents and youth health, the recent years

have seen many newly emerging issues among youth such as: stress, anxiety,

depression, homicide, mental and physical, emotional and behavioral

disorder, drug abuse Among these, adolescent violence is being given great attention in terms of protection and prevention.

During 2003-2008, the Vietnamese Ministry of Health collaborated

with the General Statistics Office (GSO), World Health Organization (WHO) and the United Nations’ Children Fund (UNICEF) to conduct two National Survey and Assessment of Vietnamese Youth (SAVY) The collected information provided many angles to the overall picture of Vietnamese youth aged 14-25 from: education, employment, sexual and reproductive health, mental health, accident, injury and diseases However, understanding the involvement of youth in violent conflict remains incomplete Despite of the fact that there have been many studies conducted on these issues by researchers such as Ha Thi Ninh and colleagues [5], Lé Thi Kim Thoa [7], Legal Department - the Vietnamese Ministry of Labour, Invalids and Social Affairs [14], comprehensive studies on youth the prevalence of violence and violent behaviors in adolescents and youth are insufficient This is

causing difficulties to the improvement of care, protection and development services for adolescents and youth.

Kinh Môn is a mountainous, rural district, used to be among poorest

and most undevelopment of Hai Duong province as well as the Red river’s

delta As growing-up of its cement industry, Kinh Môn district nowadays has

a very high speed of urbanization (the Hoang Thach cement plant in MinhTân, Kinh Môn, Hải Duong is among largest cement plants of South EastAsia) At present, there are not yet complete studies on these adolescentviolent situation and violent behavior at grassroots level This is the main

driver of our research conducted in Minh Tân town and Phú Thứ commune, Kinh Môn district, Hải Duong province which is aimed at identifying “the

violent situation among adolescents and concerned risk factors”.

The research findings are intended to imply a set of intervention measures that support to managers and policy makers to promote the prevention of adolescent violence in the coming time.

1.1 Background definitions

1.1.1 The concept of “adolescent”

In general, adolescents are a group of population made up of individuals with ages range from 10-19 [16, 25], relatively distinguished with other population groups defined by age criteria In terms of age consideration alone, adolescent can be seen as a large population group consisted of /or mixed by different age groups such as children below 16 years old [10], youngsters from 18 years old [11] or from 15 to 24 years old as defined by the

United Nations ', teenagers/ juveniles range from 14 tol8 years old [9] or

below 18 years old [11], teenager workers are those below 18 years old [8]

It can be observed that most definitions of adolescent being used today tend to mention the second decade of human life, the period in which most people have neither grown up nor childish Taking this view, our research

refer to the concept of adolescent that is to include teenagers from 13-18 years

old which is conformable with definitions of both the United Nations and

Vietnamese laws.

1.1.2 The concept of “violence”

WHO defines “violence” is “the intentional use of physical force or

power, threatened or actual, against oneself, another person, or against a

group or community, that either results in or has a high likelihood of

There is no universally agreed upon definition of youth A chronological definition of who is young, as opposed to who is a child or who is an adult, varies with each nation and culture However, the United

Nations, for statistical purposes, defines those persons between the ages of 15 and 24 as youth without

prejudice to other definitions by Member States.” (A/36/215, annex)." Page 12

http:/Awww.un.org/esa/population/publications/technicalpapers/TP2011-1.pdf

resulting in injury, death, psychological harm, maldevelopment, or

deprivation”.

From this definition, the simplest understanding of violence is use of

force, or threat of force by one to cause harms to other people or a group of people Violence takes many forms such as: brutalization, abusiveness,

harassment, torture, sex abuse, causing damages to human physical and mental condition, properties and the mankind’s progress.

Basically, we may consider violent behaviors among three groups below: violent behaviors which a person causes for his or herself (the attacker is the

victim simultaneously); interpersonal violent behaviors - or violent acts against a small group; and collective violent behaviors - often are organized violent acts of terrorists, armed troops In this study we focus on the first two groups.

1.2 Overview of the Vietnamese Government policies relating to

protection and prevention of adolescent violence

1.2.1 Adolescent care, protection, education and violent situation in

Vietnam

Since its independence in 1945, Vietnam has always paid great attention to protection and education of children, in order to provide to its

future owners a comprehensive development physically, intellectually,

spiritually and morally It can be said that up to now, in Vietnam, a complete legal framework has been created to enable every right of children (see Annex 1) With many policies and national targeted programs underway, Vietnam has significantly improved people living standards, universalized primary education, gradually indiscriminate gender imbalance and recorded many

> http://www who int/violenceprevention/approach/definition/en/index html

progresses that have been well recognized at the 10" MDG Review Summit

by the United Nations.

However, many hiden and uncovered risks are available that may cause

violence, abuse in adolescents As reported by the Ministry of Labour,

Invalids and Social affairs’, in two years of 2008-2009, there were 5,956

child-abuse cases recored, detected and processed (about 3,000 cases/year in

average) Many adolescents were treated badly, abused by their parents, relatives, teachers and guardians who should have been responsible for

nurturing and caring The report quoted sources from the Ministry of Education and Training that in school-year 2009-2010, there were 1,598 cases

of fighting among students in/outside schools recored nationwide These schools have disciplined 881 students; reprimand and warn 1,558 students; forced to be monitored at home (3 days, 1 weeks or | year) 735 students.

With the number of schools and students as the present, it is estimated that 5,260 students or 9 schools would occur a fight case; 10,000 students

would occur a reprimand case; 5,555 students would occur a warning case

caused by fighting; and 11,111 students would occur a case that force to be time-limited monitoring at home Most of these recored cases drew from minor details, disagreements, then students fought each other physically by hands, were stopped in time so there were no serious consequences Although, there were few cases that students used weapon caused serious hurts to their

adolescents must be counted firstly Because families and communities did not attach importance to the role of protection, care and education as they should be, therefore knowledge and skills on these matters of parents, guardians and adolescents themselves were also limited, many child- unsupported behaviors, habits and customs have not yet taken into attention such as teaching adolescents by beating, insulting Adolescents were vulnerable to be victims of violent behaviors, sexual abuses and crime commitment Beside of these problems, corollaries of black websites, violent games, those were out of control, also contributed a part to the violent and

abuse situation in adolescents.

1.2.2 Contemporary limitations in Vietnamese policies towards

adolescents

Limitation on the Government policies to adolescents is due to the violent and abuse situation is a new problem has been getting serious over recent years in Vietnam Therefore, it has required more attentions and policies in order to efficiently deal with adolescent violence/ abuse that is

increasingly complicated.

Many national programs to support adolescent health services have been deployed since early years of the new millennium, aiming to the outbreak of AIDS and other challenges such as traffic injuries, substance abuse (“National Action Program for Children 2001-2010”; “National Healthcare Program for mothers and children and family planning”;

“National Master plan on the protection, care and strengthen healthcare for adolescents and youth 2006-2010, vision to 2020”; “Guideline for deployment

of reproductive health services and safe sex for youngsters”, ) After all, these programs have only focused primarily on improving adolescent knowledge, competence in self-protecting and caring their health (mostly

HIV-infection, reinforcing public awareness in obeying the traffic rules in order

to reduce death toll, fatal injuries.

With_regard_to protection and prevention of youth injury issue,

Vietnam has implemented some national programs aiming at reducing fatal

injuries in youth and adolescents — mostly caused by drowning and traffic

accidents (‘The national policy on preventing accidents and injuries”; “The

National Strategy for safer transportation to 2020 with a vision to 2030”; “Building

children-friendly communities’’).

To_adolescent violence, until now, there is no comprehensive survey in

national scale According to the report “An analysis of the situation of children in Vietnam - 2010712], the issue of protection and prevention youth injuries is a new matter, its legal instruction and regulations are improving, law enforcing is not sufficient In order to temporally solve the inadequancies involving adolescent violence, the contents of violent protection and prevention are integrated into other programs such as reproductive care for female youths, domestic violent

prevention, gender quality

1,3 Modeling causes of adolescent violent behaviors

1.3.1 Causal analysis model

Understanding the root of problems that adolescents are facing is a challenging task because of its complicated nature [12] For example, from the view of policy makers/ analysts (see Figure /.3.1), while studying signs of violation of adolescent legal rights such as adolescents are under abusive, exploitation, violence, wandering or delinquency , we would consider the

reasons, high risk behaviors causing these issues.

There are many directed causes (e.g immigration, highly risk behaviors, weak protect measures, vulnerable/ malfunctioned families that

could not protect its members) contribute a large part of the situation in which adolescents are required to be special caring.

Indirect causes such as changing in social-economic elements, weak

legal framework and law enforcement would also be counted.

This is a foundation to identify the root of problems that are violating

legal rights of adolescents (poverty, inequality, gender discrimination, social

value decaying, and weak jurisdiction).

Solving the root issues often is a time-consuming task that may last over one or a few decades However, we may eliminate and control direct and indirect causes by using short-term interventions [12].

NOIL2IGSIWf AVA NOLLVAONNI SANTVA TVIDOS ALITWNOANI ALUIAOd

— ——_,

s8ue8 pozrue8o Surpnjout

'3souoatsnqp 1oj Apeat axe £ya1908 Jo syed ouuos - soSueyp tpi yeap 0198poAOtP]10 4271 ~ SuOnoUeS Jo 391 - sorjaurey djoy 01 pasoddns ame taupJT2 -

uondnusoy sofou Aqrarey wr Buen) ~ Ked 01 saotasas [EI9OS

-sromoutegy eBoy i2[đutoo v Jo 490°] - ‘Aot00s mouyo samssaud pure aouomuy - 301098 areaud Jo nun e soutosaq Aqiureg ~

AHO NVH NQO LVN dYHd IHL ONHL

1.3.2 Systematize problems by the Social-Ecological model

Apart from setting a cause analysis framework, we may systematize adolescent challenging issues by using a social-ecological model [6, 18, 20, 24] The model plans all characteristics that are identified as relating to the research problem and grouping them into levels from low to high, simple to complex These characteristics include two sets of factors those positively impact to the issue — protective factors; and those negatively impact to the issue — risk factors Studying on these factors will help to deeply understand risky behaviors On that basic, it can be easy to consider, propose an effective intervention model that focus upon mobilizing resources, dealing with the root of the matter.

For example, the case of using social-ecological model for Vietnamese

adolescent health care [6], risky behaviors to adolescent health were identified

include: (i) substance use such as tobacco, alcohol, marijuana , (ii) behaviors

relating to diet and exercise such as inactivity, unhealthy eating, eating

10

such as weapon carrying, interpersonal violence, seatbelt non use, helmet non

use, drinking and driving and sexual assaults, (iv) behaviors relating to sexualand reproductive such as non contraception, condom avoidance, early sexualdebut, multiple sexual partners, (v) social impacts such as early school

leaving, arrest.

Research on above-mentioned risk behaviors in both positive as well as

negative impacts, the model has grouped factors into levels such asindividual, peer, family, school, community (see below) to facilitate analyzing

and finding the causes for building intervention solutions.

Individual level

~ Risk factors Protective factors

isability/ chronic illness Social skills

| Intellectual impairment Average intelligence

Aggressive temperament Positive self-image

Impulsivity Perceived importance of parents

Stress

Minority status

Peer level

[ xã le ‘factors ~~ Protective factors: a

Discrimination ˆ ze Large and diverse peer network =

Bully/ harassment Low risk friends

Lack of friends Pro-social peers

Socially deviant peers

— Family level

‘Low parental education

Family mental illness

Large family size

Poverty

Exposure to family violence

Family values disruption

Family connectedness Parental monitoring

Parental expectations for schools and

behaviors

Fewer siblings Family cohession

School level

7 Risk factors ề Protective factors

Low teacher expectations Connectedness to school

Absenteeism Better grades

Abusive teachers Teacher supports

Consistency of schools attened

Community level

——

— Risk factors Protective factors

Poverty |Eduealonalaainmentbyag |

Single partents Health care facilities, utilization

Easy access to tobacco, alcohol, drugs Low rate of unemployments

Easy access to massage parlors, karaoke | Informal supports

bars Pro-social media

It is seen that, the social-ecological model is flexible and highly reliable, possible to apply in many different studies in Vietnam and over the world In this study, we expect the using of this model may help us to be

12

province from which to draw some policy implications.

1.4 Some researches on adolescent violence worldwide

Herrenkohl and his colleagues [20] used the social-ecological model to

observe on the sample of 1053 students from 10-16 years old (the SeattleSocial Development Project - 1985) in order to confirm earlier research

findings on risk factors for youth violence and to explore the effects of violent

behavior to increase risk for other problem behaviors Its result showed that at

each age, there were risk factors strongly related to later violence of

adolescents distributed among all five domains Noteworthy risk factors

included: hyper-activity, low academic performance, peer delinquency,

availability of drugs in the neighborhood.

Sutherland [23] studied the relationship of the adolescent violent

behaviors with some adolescent manifested behaviors such as alcoholconsuming, substance abusing grouped by social-demographiccharacteristics His study observed on the sample of 13.650 students from 11-

16 years old showed that: violent behaviors of adolescents did not happenaccidentally, rather it were part of a complex series of negative behaviorswhich could be named ‘delinquent behaviors” Drinking and/ or smokingadolescents involved to physical violent behaviors more frequent thanadolescents without drink alcohol and / or smoking (45% and 44% incomparison with 3,7% and 10,1% respectively) Adolescents, who used illicitdrugs, also showed a strong association with violence This study confirmedthat “rather than directly address the issue of violent behavior, researchersand violence prevention specialists would be better addressing the underlying

causes of adolescent violence”.

In 1997, Iceland conducted a national survey on violent behaviors among adolescents aged 15-16 years old [19] This study explored socio-

demographic background, social support, negative life events, psychologicaldistress, and substance use in relation to violent behaviors Its results showed

that gender, parental support, life stress, anger/ aggression and substance use

were significantly related to the perpetration of violent actions However, itdid find a clear relation between education, parental education, peer effects

and violence of Icelandic adolescents.

Shek did a research on violent behaviors in Chinese adolescents with a

link to the economic disadvantage [22] His findings showed that, among poor

adolescents, perceptions about poverty were related to violent behavior Thiswas because they believed that poverty caused by external factors so theywould displace their anger and negative emotion onto others, thus leading tomore adolescent violent behaviors He also observed that negative familyfunctioning related to poor adolescent adjustment Therefore, positive family

functioning would be considered as a protective factor for the development of

violent behavior in adolescents with economic disadvantage.

1.5 Objectives, research questions

From the above studies, we would use the social-ecological model [18,20] to exam the adolescent violence and look for related risk factors based onsurvey in Kinh Môn district, Hai Duong province However, there is adifference from Herrenkohl and his collaborators [20, 21], who used a social- ecological model to analyse risk factors of adolescent violence in all 5

domains/ levels: individual, peer, family, school and community, we onlyfocus on the first 3 domains/ levels including: individual, peer and family as

core domains/ levels that frame almost all of adolescent behaviors.

14

Adolescent Violent Situation

Related Risk Factors

| Individual domain Peer domain Family domain

- Father aleohol addict

~ Father smoke addict

- Father uses violent punishment

and discipline

- Mother alcohol addict

~ Mother smoke addict

~ Mother uses violent punishment and discipline

1.6 Research objectives and questions

1.6.1 Research objectives

Identifying risk factors related to violent behaviors of adolescents inPhú Thứ commune and Minh Tân town, Kinh Môn district, Hải Dương

province.

2.1 Research methodology

The study includes a quantitative, cross-sectional, descriptive research method with analyzing The detect-problem design is applied into the self- report questionnaire This is a suitable selection for our study because we examine the problem at a specified time.

The study’s questionnaires were created based on experience of the SAVY, round 2 This national survey conducted by the General Statistics Office and has finished its second round of consultation With the presented outcomes, its questionnaires confimed they are reliable to use Additionally,

we had refer to some other Vietnamese studies on adolescent health, run pilot questionnaires with 20 research subjects in Phú Thứ commune, with

necessary revision to be compatible with the research areas.

- Local adolescents who have been living in the research locations for

at least 03 months before our survey started, from 13 - 18 year olds;

- Living in a family that has both father and mother;

- Illiterate and/or disabilities are not eligible because the study applies

self-report questionnaires.

Secondly, we do sampling by selecting elements at regular intervals Totally 1,335 from Phú Thứ and 50! from Minh Tân adolescents are satisfied the above scheme Apply the bellow result of sample size

calculation (section 2.3.), we must select 392 subjects, so 285 must be selected in Phú Thứ with k element = 1335/285=4 and 107 subjects must be

selected from Minh Tan with k element = 501/107 = 4.

Zep = 1.96 with a liability of 95%

P: is the estimated proportion of adolescents who have violentbehaviors Because we do not have any previous research in Vietnamese

adolescent’s violent behaviors so we assume p ~ 0.5

Q=1-P

d: is expected error 5% = 0.05

Apply this formula, we have n= 196 To strengthen the sample, we

double the result so the final sample size is 392.

The study’s sample is determined randomly As the behavior of causing injury to others is a sensitive subject, probability may be low, therefore we

decided to double the sample size during the interview process in order to

ensure the accuracy of data.

The study has had a strong support from the volunteer network of Hai Duong Women’s Union during our interview and data collection so all our questionnaires were completed in time The collected data has _ been

18

sample’s results showed as follows:

at public functional houses or meeting rooms of communities) Then, we directly delivered questionnaires to subjects for filling-in by themselves.

Before subjects starting, we instructed them how to complete the survey and explained some unclear questions While subjects working on the

questionnaires, we continued to give more explanation rising by any subjects

but we did not interfere in the contents.

This self-report questionnaire approach has been used in SAVY for

sensitive contents.

2.5 Research site

The study selected one commune (Phú Thứ) and one town (Minh Tân)

of Kinh Môn district, Hải Duong province for our research.

Kinh Môn is a mountainous, rural district located in the Eastern part of Hai Duong province characterized by mountainous terrain divided by small rivers and springs and a road system less convenient than other neighbor locations This used to cause certain difficulties to the district’s socio-

economic conditions and living standards of local residents.

In the past years, together with efforts on economic development (including attracting investment to development of local industries, most typical is the Hoang Thach Cement factory - one of the largest cement

factories in South East Asia), lives of local people have been improved visually Many new urban areas have emerged in replacement of and changed the face of ruralities of Kinh Môn district.

Together with rapid economic changes, the district has faced challenges

in social orders, as it has been reported by the provincial Women’s Union, Kinh Môn district has become a hot spot of social problems, especially the

problems of HIV/AIDS, hard drugs/ alcohol addicts and violence.

Study on situation of violent adolescents in Kinh Môn district would

hopefully reveal impacts of the globalizing and integrating process to the mental and physical health development of adolescents.

2.6 Process and Analysis

We used SPSS version 16.0 for processing and analyzing data.

20

3.1 Characteristics of the research subject

The research sample has common characteristics as follows (Table

3.1.1):

- The proportion of male adolescents is higher female with 55.1% and

44.9% respectively;

- Over 3/4 of adolescents are rural residents;

- About 2/3 of adolescents are from 16-18 year olds;

- About 2/3 of adolescents are with secondary education level.

Table 3.1.1: Distribution of target group by gender, age group, resident

location and education

High school 152 38.8

_\, ,#" ASST otal ees 392 100

L a các DA Lh Vênn fe as Che 0)

3.2 Description of violent situation in the target group

3.2.1 Overview of violent situation There are 27.8% of research subjects (N=392) ever injured others 33%

male adolescents reported that they ever injured others vs 21% female

adolescents The odds ratio (OR) for male adolescents ever injured others andfemale adolescents ever injured others is 1.88, with the 95% confidence

interval (Cl) is 1.19-2.98 We would say that being male adolescents are

significantly increase risk of injuring others.

65% adolescents living in the urban area reported they ever injuredothers while those in the rural area are 14% The OR for urban adolescentsever injured others and rural adolescents ever injured others is 11.1, with the 95% Cl is 6.6-18.7 It may say that being urban adolescents are significantly

increase risk of injuring others.

Other social-demographic characteristics such as age group and education do not show relationship to violent behaviors of adolescents (ever

injured others) between the groups compared (see Table 3.2.1.1)

Table 3.2.1.1: Distribution of adolescent violent situation by gender, age

group, resident location and education

Ever injured others

OR

Never | Total 69| 38]

3.2.2 Individual domain - Individual behaviors related to adolescent

violent situation

Among potential risk behaviors of the individual domain are considered

in this study, we observed that the relationship between the factor of intended

to commit suicide and the factor of use internet and the adolescent violentsituation (ever injured others) do not have any statistically significant

However, adolescents who have one of these behaviors (maybe

combine): smoking, drinking, drunk, injured by others are significantly

increase risk of injuring others.

53.7% of smoking adolescents reported that they have ever injured

others vs 22.5% nonsmoking adolescents have ever injured others The OR for smoking adolescents ever injured others and non-smoking adolescents

ever injured others is 4.01, with the 95% CI is 2.3-6.9 It may say thatsmoking behavior is significantly increase risk of injuring others of

adolescents.

41.2% of those ever drinking and 57.7% ever been drunk reported that

they ever injured others while those without drinking and/or never been drunkyet reported injured others were 18.5% and 20.4% respectively The OR for

those groups were 3.1 and 5.3 with the 95% Cl correspondingly were 1.96-4.8and 3.1-9.0 It may say that drinking and specially to be drunk are

significantly increase risk of injuring others.

The group has ever been injured by others responded that 72.7% of

them ever injured others in comparison with 16.8% those whom have never

been injured by others The OR for risk of injuring others of the first group is13.2 times higher than the second group with 95% Cl is 7.4-23.5 (see Table

Using internet

| 56

100.0%

336 100.0%

392

100.0%

3.2.3 Peer domain - Close friend behaviors related to adolescent

violent situation

The data shows that close friends who have negative behaviors such assmoking, drinking, hard drugs using, weapon carrying, truanting do not have

a statistically significant in association with violent situation of adolescents

For adolescents whose close friends are drug addicts are not common in

the research sample (01 case) Thus although the OR for the group ofadolescents ever injured others and has close friends are drug addicts and the

group of those do not have addictive friends is 3.6, with 95% C I is 3.1-4.3,

we may see this observation likely to have a bias/error.

The interesting point here is that substance using like smoking, alcohol

consuming has a statistically significant in association with adolescent violent

behaviors but it does not spread over friends of them (see Table 3.2.3.1).

Table 3.2.3.1: Distribution of adolescent violent situation related to friends’

95% CI

100.0% 100.0%

=

108 391 27.6% 100.0%

3.2.4 Family - Family effects related to adolescent violent situation

The observation shows that the situation of abusing drugs, alcohol,smoking or applying corporal punishments do not lead to the differenceamong groups of adolescents in the risk of causing physical violence

However, family financial status of adolescent families is significantly

changed risk of the adolescent violent situation 47.2% adolescents of middle class families group reported that they ever injured others vs 19.0% of the poorer families group The OR for the risk of injuring others of adolescents in poor families lower 0.3 times than adolescents in middle class families, with

95% C.I 0.2-0.4 (see Table 3.2.4.1).

Table 3.2.4.1: Distribution of adolescent violent situation related to risk

3.3 Understanding the association between adolescent violent situation

and risk factors (logistic regression analysis)

According to theoretical research model, we are studying 22 risk

factors that relate to violent situation of adolescents as follows:

| Individual domain (10) Peer domain (5) Family domain (7) |

- Gender - Smoking - Family financial status

Ï Age group - Drinking | - Father alcohol addict

- Resident Location - Hard drug using - Father smoke addict

domain (10) Peer domain (5) Family domain (7)

-‘ducation - Weapon carrying - |~Father uses serve

- Smoking - Often truant violent punishments and

- Drinking disciplines

- Drunk - Mother alcohol addict

- Injured by other | - Mother smoke addict |

|- Intend to suicide - Mother uses serve

- Use internet violent punishments and

disciplines

However, to risk factors which have few observed cases, we do not

take into the logistic regression analyzing in order to control bias to its results.

List of unconsidered factors includes:

- Intend to suicide (Individual domain);

- Hard drug using, truanting, weapon carrying (Peer domain);

- Mother - alcohol/smoke addict (family domain).

Global test of parameters of the regression model showed that Sig

<0,005, means there are relationship between considering risk factors with

adolescent violent situation in the research areas (Table 3.3.1.).

Table 3.3.1.: Omnibus Tests of Model Coefficients

eS eS

Step 1 Step 155.301 16 000

155.301 16 000 155.301 16 000

30

Block

Model

= 0.472 shows about 47.2% cases is explained by this logistic model Withthe sensitive matter such as the adolescent violent situation, we may see this

Cox & Snell R

-2 Log likelihood Nagelkerke R Square

Square

Hosmer and Lemeshow Test of goodness of fit with Sig.<0,001 shows

that the observed and predicted probabilities match (Table 3.3.3).

Table 3.3.3.: Hosmer and Lemeshow Test

The results show that the overall accuracy of this model to predict

violent behavior is 85.5% (with a predicted probability of 0.5 or greater) Thesensitivity is given by 68/109 = 62.4% and the specificity is 267/283 = 94.3%.Positive predictive value = 68/84 = 80.1% and negative predictive value =

267/308 =86.7% (Table 3.3.4).

Table 3.3.4.: Classification table

Predicted Ever injured others

% correct