(PDF) Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

Study on

Gender, Masculinity and Son Preference

in

Nepal and Vietnam

Contributing Authors

ICRW

Dr. Priya Nanda

Mr. Abhishek Gautam

Dr. Ravi Verma

ISDS

Dr. Khuat Thu Hong

Ms. Tran Giang Linh

CREHPA

Dr. Mahesh Puri

Ms. Jyotsna Tamang

Mr. Prabhat Lamichhane

This report provides the results of the study undertaken by International Center for Research on

Women (ICRW) in partnership with Center for Research on Environment, Health and Population

Activities (CREHPA) in Nepal and Institute for Social Development Studies (ISDS), Vietnam. The

study was commissioned by UNFPA, Asia and Paci c Regional Of ce, Bangkok and funded by

Australian Government, AusAID. The objective of the study was to explore the gender norms,

masculine behavior and attitude towards son preference in Nepal and Vietnam.

Disclaimer

Support for the study was provided by Australian Government, AusAID through UNFPA, Asia

and Paci c Regional Of ce – the views in this report are not necessary those of AusAID.

Suggested Citation

Nanda Priya, Gautam Abhishek, Verma Ravi, Hong Khuat Thu, Puri Mahesh, Linh Tran Giang,

Tamang Jyotsna, Lamichhane Prabhat (2012). “Study on Gender, Masculinity and Son Preference

in Nepal and Vietnam”. New Delhi, International Center for Research on Women

This document may be reproduced in whole or in part without permission of the International

Center for Research on Women (ICRW) provided full source citation is given and the

reproductions is not for commercial purposes.

ACKNOWLEDGEMENTS

We are thankful to UNFPA Asia Paci c Regional Of ce for commissioning the study on Gender,

Masculinity and Son Preference, and are grateful to the Australian Government, AusAID for

their generous support for these two pioneering national studies and this publication. We are

particularly thankful to Dr. Anand Tamang, Director, Center for Research on Environment, Health

and Population Activities (CREHPA), Nepal and the team, Dr. Le Bach Duong, Director, Institute

for Social Development Studies (ISDS), Vietnam and the team, for their valuable partnership in

conducting the study in their respective countries.

We sincerely thank Ms. Kiran Bhatia, Gender Advisor, UNFPA Asia and Paci c Regional Of ce

Bangkok, for the conceptual development of research on this unexplored domain of men and

son preference and the lead technical oversight and guidance she provided throughout the

study. We would also like to acknowledge the valuable support provided by Galanne Deressa

Programme Specialist and Patnarin Sutthirak, Programme Associate at UNFPA APRO. We would

like to extend our gratitude to Ms. Nobuko Horibe, Director, UNFPA Asia and Paci c Regional

Of ce Bangkok for their support to the study. We would like to thank Mr. Bruce Campbell

and his team at UNFPA Country Of ce Vietnam for their technical and  nancial support to the

study in Vietnam.

We are also thankful to Dr. K.M. Sathyanarayana and Dr. Sanjay Kumar from, UNFPA India of ce

for their inputs during  nalizing of the study design. We are particularly thankful to Professor

Christophe Z. Guilmoto from Centre for Population and Development, France and consultant for

UNFPA, Emma Fullu, Research Specialist and James L. Lang, Program Coordinator from Partner

for Prevention (P4P), Bangkok for their suggestions on the study population and instruments.

The team would also like to extended heartfelt thanks to Dr. Ajay Kumar Singh, former Technical

Specialist at ICRW, ARO and Ms. Sonvi Kapoor, former Research Associate at ICRW, ARO, for

their contributions in  nalizing the study instruments, sampling design and guiding the teams

in Nepal and Vietnam during data collection and analysis. We could not have managed the

study execution without their contributions. We also thank Ms. Anuradha Bhasin, Consultant at

ICRW for her contribution during the preparation of the study report. We extend our thanks to

Dr. Ellen Weiss, Senior Advisor, ICRW, Head Quarters (HQ) for her editorial support in  nalizing

the report. We are also thankful to Caroline Klein, Budget and Sub-grant Manager at ICRW,

HQ and Sandeepa Fanda, ICRW, ARO for their invaluable administrative support. We would

like to thank Ms Chandana Anusha who also contributed in the tool development process as

a research fellow with ICRW.

We would like to express our heartfelt thanks to the Institutional Review Board Committees of

the International Center for Research on Women and the Center for Research on Environment,

Health and Population Activities for providing the necessary ethical approvals to carry out the

study.

The contributions of the study participants who patiently responded to questions on personal

matters, included in the study questionnaire, are gratefully acknowledged. The study would

not have been possible without their voluntary and enthusiastic participation.

Acronyms

Executive Summary

1

Chapter 1: Introduction 5

1.1 Background 5

1.2 Research Objectives 6

1.3 Study Context 6

1.3.1 Nepal 6

1.3.2 Vietnam 8

Chapter 2: Methodology 11

2.1 Conceptual Framework 11

2.2 Sampling Design and Sample Size 12

2.3 Study Tools 13

2.4 Conducting the Survey 14

2.5 Analysis and Variables 15

2.6 Ethical Considerations 16

2.7 Challenges and Limitations of the Data 16

Chapter 3: Socio-Demographic Profile of Respondents 17

3.1 Demographic Characteristics 17

3.2 Socio-economic Characteristics 18

3.3 Economic Stress 21

3.4 Partners Characteristics 22

3.5 Alcohol and Substance Use 22

3.6 Other Indicators 23

Chapter 4: Men’s Attitude towards Gender Equality 25

4.1 Men’s Attitude towards Gender Equality – 25

The Gender Equitable Men (GEM) Scale

4.2 Factors associated with Men’s Attitude towards Gender Equality 28

4.3 Childhood Experience of Gender Inequality 29

4.4 Factors Associated with Childhood Gender Inequality 30

4.5 Participation in Maternal Health Care and Child Care 32

4.6 Men’s Participation in Domestic Duties 34

CONTENTS

Chapter 5: Men’s Attitude towards and Experience of Violence 35

5.1 Type of Violence Reported Against Wife’s/Partners 35

5.2 Violence by Selected Socio-Demographic Characteristics 37

5.3 Perpetration of Violence by Selected Background Characteristics 39

5.4 Factors Associated with Lifetime Violence 40

Chapter 6: Men’s Attitude towards Son Preference 44

6.1 Son Preference Attitude 44

6.2 Factors Associated with a Son Preference Attitude 46

6.3 Perceived Importance of Having a Girl or a Boy Child 51

Chapter 7: Men’s Knowledge about and Attitude towards Laws and Policies on 54

Reproductive Health and Rights

7.1 Knowledge about Abortion Laws and Services 54

7.2 Attitude towards Abortion (By different Socio-Demographic Factors) 56

7.3 Knowledge about Ultra Sound Test by Wife/Partner and 57

Attitude towards Sex Selective Abortion

7.4 Men’s Knowledge and Attitude towards Policies and 59

Laws Promoting Gender Equity

7.5 Knowledge and Attitude towards Inheritance Rights 61

7.6 Knowledge and Attitude towards Laws on Violence Against Women 62

7.7 Factors Associated with Knowledge about Different Gender Related Laws 63

Chapter 8: Conclusion and Recommendations 65

Annexure: Tables and Figures 69

References 84

CBS Central Bureau of Statistics

CREHPA Center for Research on Environment, Health and Population Activities

DHS District Health Survey

DFID Department for International Development

FWLD Forum for Women Law and Development

GBV Gender Based Violence

GDP Gross Domestic Product

GEM Gender-Equitable Men

GoN Government of Nepal

GSO General Statistics Office

HIV Human Immunodeficiency Virus

ICRW International Center for Research on Women

IMAGES International Men and Gender Equality Survey

IPV Intimate Partner Violence

IRB Institutional Review Board

IRC Institutional Review Committee

PSU Primary Sampling Unit

P4P Partners for Prevention

PATH Program for Appropriate Technology in Health

PPS Probability Proportionate to Size

SLC School Leaving Certificate

SPSS Statistical Package for Social Science

SRU1 Sex Ratio Under age 1

STI Sexually Transmitted Infections

UNDP United Nations Development Program

UNESCO United Nations Educational, Scientific and Cultural Organization

UNFPA United Nations Population Fund

VAW Violence against Women

VDC Village Development Committee

WHO World Health Organization

LIST OF ACRONYMS

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 1

Strong son preference in parts of Asia has led to serious discriminatory practices towards girls

and women, with disturbing effects on their status, health and development. The significant

male surplus in some populations resulting from the excess of male births since 1980 has

an impact on the sex ratios at birth and subsequently affected the dynamics of marriage

in several regions of this continent. The lack of women available for marriage has created

further discrimination through increased violence against women, trafficking, abduction, forced

marriages, or sharing brides among brothers as different practices in different countries. Girls

who are born also face discrimination in their lives through unequal opportunities for health

care, education and fulfilling their aspirations.

This study on Men’s Attitudes towards Son Preference has arisen out of such concerns that

have been observed in several Asian countries now for decades. This study was conceptualized

and designed by the International Center for Research on Women (ICRW) in New Delhi and

was implemented in partnership with two research partners in Nepal and Vietnam. In Nepal

the study partner was Center for Research on Environment Health and Population Activities

(CREHPA) and in Vietnam, the Institute for Social Development Studies (ISDS). The overall

objective of this study is to understand the dimensions, nature and determinants of men’s

varying attitudes to son preference and gender-based violence. The study adapted the

International Men and Gender Equality Survey (IMAGES) survey tool to explore men’s attitude

towards son preference. IMAGES is one of the most comprehensive surveys ever carried out

on the attitudes and behaviors of men aged 18-49 years, on issues related to gender equality,

including sexual and reproductive health, maternal health, gender-based violence and men’s

participation in care-giving and family life.

The household survey with men in both countries was carried out in July-August, 2011. In

Nepal, total sample of 1000 men in the age group 18-49 were interviewed across three districts

namely Dang, Gorkha and Saptari; in Vietnam the sample comprised of 1424 men from two

districts namely Hung Yen in North and Can Tho in south.

The mean age of the men participating in the survey was 32 and 35 in Nepal and Vietnam

respectively. Three quarters of men in both the countries reported to be married and one-

third of the unmarried men were cohabiting. Level of education was high in Vietnam, among

the surveyed men. All of them were literate and only two percent of men reported that they

did not attend any formal education. In Nepal, eight percent men were illiterate while others

had been to school and one-fifth of them had attended higher secondary. Nepal being a

Hindu dominated country majority of men their followed the Hindu religion while in Vietnam

little less than two-thirds reported to follow no religion, followed by fifteen percent following

Christianity. In terms of occupation, in Nepal nearly half were farmers followed by nearly one-

fourth of men who were in service. In Vietnam, around one-third men reported farming as an

occupation and nearly one-fourth were working as manual labor.

EXECUTIVE SUMMARY

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

2

The study shows that in both the countries majority of men had a moderately gender equitable

(neither high nor low) attitude but they favor a conservative gender roles for women. In Nepal,

nearly half of the men and in Vietnam more than three-quarter males agreed that a woman’s

important role is to take care of her home and cook for her family. Interestingly in both

countries there was very less agreement on the statement that a woman’s most important

role is to produce a son for her husband’s family. On attitudes around violence against women,

in Nepal 44 percent men agree that a woman deserves to be beaten while in Vietnam it was

26 percent. Norms around masculinity was found to be high in both countries; in Vietnam

90 percent men agreed that to be a man you need to be tough. In Nepal 70 percent of men

agreed on the same statement. On the value of sons versus daughters, men’s attitudes were

similar in both countries. Majority (90 percent) of men gave agreement to the statements like

‘a man with only daughters is unfortunate’ and ‘not having a son reflects bad karma and lack

of moral virtue’.

Education, occupation and religion were found to be associated with men’s attitude towards

gender equitable norms in both countries. Men with higher education, working as professionals

are more gender equitable.

Men in Nepal and Vietnam are brought up in a socio-familial context where gender discrimination

against women is common. More than half (55 percent) of the men in Nepal and two-thirds

(66 percent) in Vietnam reported experiencing or witnessing the gender inequality in childhood

and the most common form witnessed was restrictions on freedom off their sisters or female

cousins. In both countries the GEM scale score was found to be significantly associated with

the experience/witness of childhood gender inequality. Other factors associated were different

in both countries. For example, caste/ethnicity and economic stress was found significantly

associated in Nepal while in Vietnam, age, education and occupation were significantly

associated.

There appears to be a clear relationship between men’s participation in maternal health care

and child care and socio-demographic characteristics. Men in both countries who accompanied

their wives/partners on their prenatal visits were more likely to be younger, live in urban

areas, have a higher level of education and be professionals. Unsurprisingly, in both countries

there was a significant relationship between men’s daily participation in the care of children

and their GEM scale scores. In addition, in Nepal, a significantly larger proportion of men in

nuclear families and who worked in farming helped with the daily care of children, while in

Vietnam, men’s level of education had a significant relationship with their involvement in the

daily care of children.

The study found a high prevalence of violence against intimate partners among surveyed men.

In Nepal, about 71 percent of men reported ever using any form of violence against their

intimate partner, while in Vietnam, the proportion was 60 percent. The most common form of

intimate partner violence (IPV) in both countries was emotional violence, which almost half the

sampled men in both countries reported having enacted at some time. Followed by emotional

violence was physical violence, reported by two-fifth of the men in Nepal and one-third of

men in Vietnam. The responses to violence against women in the past year showed that more

than 40 percent of men in Nepal and 25 percent of the Vietnamese men had committed some

form of violence in the past year. In both countries, the most common form of violence in the

last year, as is the case with ever use, was emotional violence, followed by physical violence.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 3

Age and occupation was significantly associated with IPV in both countries. Older men were

more likely to perpetrate violence as compared to men in the age groups 18-24 years in

both countries. Occupation levels too have significant effects, men in business or working

in shops had a higher likelihood of perpetrating violence as compared to men who were

professionals. Education too showed an association in the expected direction but was not

statistically significant. Importantly, men with high gender equitable attitude were less likely

to perpetrate violence. Unsurprisingly, childhood experiences of bullying and gender inequality

played a strong role in both countries as men with experiences of either were almost twice as

likely to engage in IPV, especially in Nepal. Similarly, the use of alcohol increased the likelihood

of IPV in both countries.

The data shows that in both countries a preference for sons exists among the surveyed men.

Most men in both countries supported statements that demonstrated son preference i.e.

related to the direct value of having sons. The highest proportion of men in both countries

agreed with the statements that sons are important to carry on the family lineage and for

support in the old age. Surprisingly, fewer men agreed with the views related to aborting a

female fetus, putting a daughter up for adoption, or abandoning a wife who does not bear

sons. In both countries, the level of education and type of occupation of men were significantly

associated with their son preference attitude. There was significant association between men’s

son preference and their gender equitable attitudes (GEM scale) and men’s control over their

wives (relationship control index) in both countries.

The patriarchal nature of Nepalese society combined with socio-economic and religious values

creates a strong influence to have a male child in the family. Similarly in Vietnam, it is the

patrilineal and patrilocal kinship system that tends to place strong normative pressure on

couples to produce at least one son. Men’s views about the importance of sons and daughters

are highly influenced by traditional customs, gender roles and expectations (eg. only boys can

carry on their father’s name and continue the family lineage while girls provide emotional

support and are expected to be dutiful and hardworking).

Men in this survey had a relatively good awareness of laws and policies on violence against

women and law against abortion. Majority of men in Nepal were not aware of correct legal

conditions for abortion and in Vietnam this percentage was low. In both countries men favored

the law on sex-selection and were aware of it. Nearly one-third of men believed that the sex-

selection law can go against women’s abortion rights and their rights to choice. On the law

regarding inheritance right, a very high percentage of men in Nepal were aware of it but in

Vietnam only one-third of men were aware of it. But, in both countries among those who

were aware, they considered these laws to be favorable.

This study was first of its kind in both countries that explored men’s attitudes on a wide

range of issues related to gender equality, son preference, the levels and types of intimate

partner violence and knowledge and attitude toward laws and policies related to women’s

right. The study affirms that high son preference, conservative gender roles and inequitable

attitudes persist in both the countries. There is need for more comprehensive, long-term and

male-targeted intervention programs or campaigns at national and local levels taking into

account men’s specific socio-economic conditions that influence their preferences. While the

levels of knowledge of laws and policies relating to gender equality is high, there is clearly a

greater need for enforcement coupled with effective messaging that not only enlightens about

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

4

the laws but also the norms underlying the practices that are being legislated. Because of the

relationship between childhood experiences of inequality, gender equitable attitudes and son

preference and IPV, early childhood interventions around masculinity and men’s role in family

also standout as policy recommendations from this study.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 5

1. INTRODUCTION

1.1 Background

Son preference is a growing problem in a number of Asian countries. Its high prevalence has

resulted in substantial levels of excess female child mortality, and has skewed sex ratios in

parts of China, India and Vietnam, with emerging signs in Nepal and Pakistan. Responsibility

for economic, socio-cultural and religious functions is vested in men in most Asian patriarchal

societies. Sons are therefore seen as essential for the survival of the family, for social security

in old age, and are assigned a greater value than daughters; they consequently have more

power and control over resources, particularly land and property, and over the women of the

family. At the state level, this power is often reflected in a country’s laws and policies, and

contributes to placing women in subordinate positions, both in the private sphere where they

are dependent on men economically, and publicly where they have little or no decision-making

power and are seen as a burden (Murphy, 2003; Das Gupta et al., 2003; Chow and Berheide,

2004; all cited in Li, 2007).

The societal importance given to boys has translated into deep-rooted discriminatory practices

toward girls and women, with devastating effects on their status, health and development and

an enormous pressure to produce sons. In the context of declining family size and restrictive

policies on reproduction and access to unregulated health services, this pressure can have

severe consequences on women’s mental and physical health. For men and boys, it has resulted

in stereo-typical perceptions about masculinity and impunity by some to practice violent

behavior that is often socially sanctioned. The significant male surplus in some populations

resulting from the excess of male births since 1980 inevitably has an impact on the context

of both women’s and men’s entry into partnership or marriage. There is evidence that the

lack of women available for marriage has led to increased violence against women, trafficking,

abduction, forced marriages or brides being shared among brothers (Guilmoto, 2007).

International initiatives to achieve desired reproductive health outcomes – such as reducing

unintended pregnancy, stopping the spread of HIV and improving maternal health are

increasingly recognizing that these outcomes are affected by gender relations, norms and roles

commonly ascribed to women and men, and associated inequalities. In response, governments

and international donor agencies have embraced the idea that reproductive health policies and

programs should support women’s empowerment and gender equality and they have included

it in their goals and strategies.

To help move forward the reproductive health and women’s empowerment agenda, rigorous

data are needed on men’s gender-related attitudes and behaviors, including around son

preference. This study contributes to the growing evidence base on men by adopting the

survey tools of the International Men and Gender Equality Survey (IMAGES) in Vietnam and

Nepal. IMAGES is one of the most comprehensive surveys ever carried out on the attitudes

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

6

and behaviors of men aged 18-49 years, on issues related to gender equality, including

sexual and reproductive health, maternal and child health, gender-based violence and men’s

participation in care-giving and family life (Barker et al, 2011). This study uses a modified

version of IMAGES to include a focus on son preference.

1.2 Research objectives

The overall objective of this study is to understand the dimensions, nature and determinants

of Nepali and Vietnamese men’s attitudes to son preference and gender-based violence. The

specific objectives of the project are to:

1) Assess men’s current behaviors and attitudes on a wide range of issues as they relate

to gender equality

2) Assess men’s knowledge and attitudes toward son preference and violence

3) Explore contributing factors that can be attributed to men’s attitudes and behaviors

related to gender-based violence and son preference

4) Assess men’s knowledge of and attitudes toward policies that have sought to promote

gender equality (e.g., gender-based violence, sex selection, family planning and inheritance

rights)

1.3 Study context: Nepal and Vietnam

1.3.1 Nepal

Nepal’s population is about 27 million with a population growth rate of 1.4 percent (CBS, 2011).

The 2001 Population census listed 103 diverse ethnic/caste groups, each with its own distinct

languages and culture. Nepal is primarily a Hindu country with more than 81 percent of its

population believing in the Hindu religion. The recent DHS study 2011 found that the total

fertility rate is 2.6 per woman, reduced from 4.1 in 2001. The literacy rate is 54 percent overall,

with an enormous gender gap (65% among men and 43% among women) (Ministry of Health

(Nepal)/New Era/ORC Macro, 2006). A large percentage of the population lives in remote areas,

without access to basic infrastructure or services. The country is divided into three geographic

regions: the terai (plains area), the hills and the mountains. As one moves from the terai up

to the mountains, living conditions and access to health care become increasingly difficult. As

a result, there are wide discrepancies in health services in different regions.

The economy is agrarian, although most households are not self-sufficient and rely on some

non-agricultural sources of revenue. Per capita GDP is estimated to be less than US$ 500;

about 25 percent of the people live below the poverty line and is ranked 157th in the 2011

Human Development Index, lowest in the South Asia region (UNDP, 2011). Development

strategies have been hindered, in part, by topography, by marked caste and class distinction

and unequal distribution of power and resources, as well as by severe gender discrimination

in spheres of public and private life.

Various forms of Gender Based Violence (GBV) exist in Nepal, including, domestic violence,

family abuse (polygamy, child marriage, dowry related violence, mental abuse), sexual violence

(trafficking, forced sex, sexual harassment), and punishment for witchcraft

(Boxi)

. The context

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 7

of GBV is interlinked with underlying social, cultural, religious and gender norms and with

political conflict in Nepal. Traditional practices also contribute to the exploitation of women,

particularly sexual abuse and sex work. For example, within the

Badi

community (in mid-

western Terai district) many women have been forced into commercial sex work. The

Deuki

tradition involves families offering young girls to temples to act as dancers in ceremonies;

however girls often experience sexual exploitation and enter sex work to earn a living. Similarly,

the

Jhuma

are Sherpa traditionally send second-born daughters to monasteries as an offering

to ensure the well-being of the girls’ family. Dalit women (lowest social caste) face multiple

discriminations, maybe accused of

Boxi

(witchcraft) and are extremely vulnerable to sexual

exploitation (Hasselman et al, 2006).

Nepal has been classified as having considerable levels of son preference since the World

Fertility Surveys first documented the phenomenon in the 1980s (Cleland et. al, 1983). It is

believed that there are certain communities which do not rejoice the birth of a daughter,

resulting in a deficit of women in those areas (World Bank/DFID, 2006). Son preference is the

direct result of the discrimination of women in all aspects of family and community life. There

are several rites and rituals that can only be performed by sons such as lighting the funeral

pyre. Economically, daughters are considered a liability because of dowry and because they

live their natal home upon marriage, so for old age security, parents can only rely economically

on sons. Socially, preference is given to sons because of their role in continuing the family

lineage. Nepal’s patrilineal social structure discourages women from practicing contraception

until they have a son. Analysis of the 1996 Nepal Demographic and Health Survey data of

5,902 women showed that the overall contraceptive prevalence was 25 percent, compared

with 33 percent if there had been no sex preference-a reduction of eight percentage points,

or 24 percent (Leone et al, 2003).

A recent study found that in both the 1991 and 2001 population censuses, the sex ratio

before one year of age is lower for females, especially in seven terai districts. The sex ratio is

higher among terai high and middle caste/ethnic groups than the hill ethnicity/caste, further

indicating ‘missing women’ in the terai belt of the country. The same study pointed to the

existence of sex selective abortion in urban areas of the terai region, where the status of

women is low and the girl child is often neglected in the family (CREHPA/UNFPA, 2007a).

Until 2002, abortion was illegal in almost all circumstances in Nepal. However, research has

found that unsafe abortions carried out clandestinely before this period were fairly widespread,

and were contributing significantly to Nepal’s high maternal mortality rates. The open border

with India, coupled with the geographical accessibility to and socio-cultural similarities of terai

inhabitants with those of Northern India, meant that it was not uncommon for women in

these areas to visit Indian border towns for health care, including abortions, which are legal

for upto 20 weeks CREHPA/UNFPA, 2007b).

The 2002 law that legalized abortion in Nepal expressly prohibits sex determination and sex-

selective abortion. A woman can legally obtain an abortion up to 12 weeks’ gestation, up to

18 weeks in case of rape or incest, and at anytime during pregnancy if her life is at risk or

the fetus has congenital anomalies. However, over the last few years, especially after the

legislation of abortion, concern has been raised about sex selective abortion in Nepal. A study

indicates that because of the legalization of abortion, the easy availability of prenatal sex-

determination technologies and abortion clinics, as well as the value given to sons, the demand

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

8

for sex-selective abortion could increase in the coming years. Another study conducted in

2010 showed that about 11 percent of Nepali women residing in border areas visited India for

sex selective abortion (CREHPA, 2010).

Within the political and legal spheres, there is widespread discrimination against Nepalese

women and girls. According to a 2009 report, “Discriminatory Laws in Nepal and their impact

on women”, Nepal still had 96 discriminatory provisions and 92 schedules in various acts

and provisions, including the Constitution, that have discriminatory provisions which provides

various rights and responsibilities only to men which indirectly encourages son preferences.

Substantive discrimination exists in the field of nationality, marriage and family relations,

sexual offence and property rights. (FWD, 2009).

Despite the many barriers, Nepal has made progress in reducing gender-based discrimination

and empowering women over the last decade. Women now have significantly better access

to education. Gender-inclusive/friendly policies and national plans of action have been framed

to increase women’s access to political, economic and social spheres, and to reduce gender-

based violence. Legal reforms have been introduced and institutional mechanisms have been

established and strengthened in order to ensure gender equality and women’s empowerment.

In addition to the legalisation of abortion, the Domestic Violence and Punishment Act was

passed in May 2009 in Nepal. The country’s gender empowerment measurement has improved

significantly from 0.391 in 2001 to 0.496 in 2006 (UNFPA, 2007).

1.3.2 Vietnam

Vietnam, with a total population of 87 million people, is ranked thirteenth in the world and

has the second largest population in South East Asia. About 70% of the population lives in

rural areas, mainly doing agricultural work and having limited access to basic infrastructure

or services. The annual population growth rate, as of 2010, is 1.05. The total fertility rate is

currently 2.00 per woman, less than 2.25 in 2001 (GSO, 2011). The contraceptive prevalence

rate was 79.5% among women aged 15-49 in 2008. The adult literacy rate was 83.7% in 2009,

with a huge gender gap, 79.2% among females and 88.3% among males (WB, 2011b).

There are more than 54 culturally distinct ethnic groups of which the Kinh people is the majority,

accounting for 89% of the total population. About 6.8 million (7.9% of the total population) are

practicing Buddhists, 5.7 million (6.6%) are Catholics, 1.4 million (1.7%) are adherents of Hoa

Hao, 0.8 million (0.9%) practice Cao Dai, and 0.7 million (0.9%) are Protestants. The vast majority

of Vietnamese people practice ancestor worship in some form (GSO, 2010). Vietnam’s average

per capita GDP has increased dramatically, from below USD 200 in 1989 to above USD 1,224

in 2010. Vietnam is ranked one of the best performing economies in the world over the last

decade and has entered the rank of middle income. Agriculture is still the cornerstone of

Vietnam’s economic productivity, accounting for almost one fifth of GDP in 2010.

Vietnam, like many countries, has high levels of violence against women. The most recent

nationally representative study on domestic violence, conducted in 2009, indicated that 58.3%

of ever-partnered women had experienced at least one of three types of domestic violence

(physical, sexual and emotional violence), at some point in their marriage. Emotional abuse is

the most frequently reported type of violence, with 54% of ever-partnered women having ever

experienced it, followed by physical abuse, with a prevalence rate during their lifetime of 32%.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 9

Sexual abuse is reported by 10% of the women surveyed (GSO, 2010). The causes of domestic

violence against women are rooted in Vietnam’s social, cultural, economic and political context.

In the Confucian tradition, women’s duties are primarily bounded with housework, procreation

and care-giving for the family members. Women are expected to endure difficulties and please

their husbands in any circumstances. Domestic violence has been seen as a sensitive and

private issue in Vietnam; thus women cannot share their experience with anyone. In the

meantime, men have an entitlement to ‘teach’ their wives in order to protect the honor of the

family as well as to show their masculinity (RydstrØm, 2006). Alcohol and a hot temper are

often used to justify men’s use of violence against women (Mai et al., 2004).

Strong son preference in Vietnam is rooted in the patrilineal and patrilocal kinship system

that tends to place strong normative pressure on people to produce at least one son. This

son preference has been perpetuated in the context of a two-child policy. Traditionally, sons

are essential to carry on family lines and names, to perform ancestor worship and to take

care of parents in their old age. Also, sons are more desired because having a son helps

improve a woman’s status within the family and a man’s masculinity and reputation within

the community. Men and women with no sons are often exposed to strong pressures from

the extended family as well as experience humiliation within the community.

In parts of Vietnam, there is an imbalance in the sex ratio at birth, which is seen as a

demographic manifestation of gender inequality corresponding discrimination against females

before birth. In order to meet their strong preference of sons, many couples tend to use

advancing technologies, including ultrasonography to determine the sex of the foetus and

induced abortion to eliminate unwanted female fetuses (UNFPA, 2011). It has been argued

that despite impressive social achievements in the improvement of women’s well-being during

the past few decades, son preference still persists and hinders the country’s efforts towards

gender equality.

In recent years, the government of Vietnam has demonstrated its strong commitment to

promote gender equality and to eliminate violence against women. The Convention on the

Elimination of Discrimination Against Women (CEDAW) was ratified by the Government in

early 1982. More importantly, in 2006, Vietnam passed the Law on Gender Equity (GEL), which

ensures gender equity in all aspects of life and assigns detailed responsibilities to specific

organizations, institutions, families and individuals in ensuring these principles. Following this,

in 2007, the Government passed the Law on Domestic Violence Prevention and Control (DVL),

which provides explicit protection from violence within the family to its members and covers

a wide range of acts of domestic violence. The Government has also produced a number

of decrees, circulars and national plans of action to address the prevention and control of

domestic violence. Reducing gender-based violence is one of the key objectives put forth by

the draft National Strategy on Gender Equality 2011-2020.

Also, since 2000, Vietnam has issued a number of policies and laws to regulate sex selection.

According to Government Decree No. 104/203/ND-CP issued in 2003, the use of ultrasound

and abortion for the purposes of sex selection is illegal. Moreover, the Decree No.114/2006/

ND-CP in 2006 also imposes financial penalties for those using ultrasound and abortion for sex

identification and revoke practicing licenses and certificates of any individual or organization

who violates the above rules for one to three months. However, in reality, the enforcement of

these decrees is not effective: there is no monitoring of health providers and no sanctions for

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

10

violations of the ban (UNFPA, 2011). Most women know the sex of their child-to-be prior to its

birth. The 2010 Population change and Family planning Survey has shown that 75.2 percent

of women aged 15-49 who gave the birth from April 2008 to March 2010, knew the sex of the

foetus before delivery. A recent study on induced abortion among women in Hanoi indicates

that women with more children, particularly those with more daughters or without a son,

were more likely to undergo a second-term abortion than a first-term abortion. The authors

estimate that, in 2003, 2 percent of all abortions to women with at least one living child were

intended to avoid the birth of a female (Belanger and Khuat, 2009).

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 11

2.1 Conceptual framework

Gender refers to the widely shared expectations and norms within a society about appropriate

male and female roles, responsibilities and behaviors, and the ways in which women and

men interact with each other (Gupta, 2000). Gender therefore encompasses masculinities and

femininities, power relations between women and men, and structural contexts that reinforce

and create these power relations. This study is framed within this understanding of gender

as relational and structural, and within the field of “masculinities,’ which seeks to understand

how men are socialized, how men’s roles are socially constructed, and how these roles and

power dynamics change over the lifecycle and in different social contexts (Connell, 1994). This

concept of masculinities also enjoins us to examine the diversity of men, the pressures they

may perceive to adhere to specific versions of manhood – particularly the widespread belief that

being a man means being a provider or having stable employment – and to understand how

these roles and power dynamics change historically and by social context within the various

domains where social meanings of gender and gender-based inequalities are constructed,

and how these are associated with son preference and intimate partner violence (Barker

et al, 2011). It is recognized that factors explored in this study may not adequately cover

all potential underlying factors linked to son preference and intimate partner violence. The

framework posits son preference attitude and use of violence as functions of some contextual

and community factors and some household, individual and interpersonal factors. Clearly, the

domains overlap to a certain extent and reciprocally influence each other.

2. METHODOLOGY

CONCEPTUAL FRAMEWORK

FIGURE 2.1

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

12

2.2. Sampling design and sample size

The study adopted a standard multi-stage cluster sampling design in both countries. The broad

approach was to first select the district/province based on the sex ratio at birth. Within the

district/province, samples were then distributed in order to get equal representation from the

rural and urban areas. The detailed sampling approach adopted in each country is elaborated

below:

In Vietnam, the survey used a multistage random sampling strategy, with clusters (enumeration

areas) as the primary sampling unit (figure A2.1). The samples were designed to be self-

weighting.

!

Stage 1: Selection of province

: The provinces of Hung Yen (sex ratio at birth, SRB = 124)

and Can Tho (SRB = 110) were selected for the survey, because of their elevated SRB

levels.

!

Stage 2: Selection of communes within each province

: Since the population sizes of Hung

Yen and Can Tho are almost the same (1,128,702 and 1,187,089, respectively), the total

study sample of 1,680 men was divided equally between the two sites. Based on the

rural/urban population distribution in Vietnam, of the 840 men in each site, 588 men

(70%) were selected from four rural communes and 252 men (30%) from four urban wards.

The eight communes were selected from each province using Probability Proportionate to

Size (PPS), based on a list of all the communes in the selected province and their sizes

(i.e., the number of households or of people).

!

Stage 3: Selection of clusters within each commune

: Using PPS, two clusters were chosen

from each selected commune.

!

Stage 4: Selection of households within a cluster:

An accurate list of households was

developed in each cluster with the assistance of local authorities and population

motivators. A random selection was made in the list to get a total of 74 households

in each cluster of each rural commune and 32 households in each cluster of each urban

ward.

!

Stage 5: Selection of men within each household:

If more than one eligible respondent

was found in the selected household, then the Kish table in the household selection form

was used to identify the eligible man from each selected household for interview (figure

A2.1).

In Nepal, the study was based on a stratified two-staged cluster sampling technique. It was

carried out in three districts (out of 75), chosen to represent consistently high, medium and

average sex ratios of children less than a year of age (SRU1), based on the 2001 population

census. It was assumed that the selection of study sites representing districts with varying

sex ratios would provide a balanced picture of men’s attitudes and behavior pertaining to son

preference and masculinity. The three districts chosen for the present study were: Saptari

(109/100 females); Gorkha (106/100 females); and Dang (102/100 females).

First stage

: The study selected 40 primary sampling units (PSUs) (24 rural wards and 16 urban

wards/sub wards) using probability proportion to size (PPS). Given the low level of urbanization

in Nepal (15 percent in the 2001 Census), the allocation of proportional urban samples would

have been inadequate to reveal significant rural and urban differences, if any. Hence, it was

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 13

decided to select 40 percent of the clusters (16 clusters) from the urban areas. To select the

clusters, VDCs (Village Development Committee) were listed alphabetically, separately for rural

and urban, with their corresponding wards/cluster and their households and populations, and

systematic random sampling was applied.

Second stage

: A sketch map of each sampled cluster was prepared and divided into 2-5

segments (in the case of more than 100 households), from which a segment was chosen

randomly. From an up-to-date list of households, 25 households were selected through

systematic random sampling. A screening questionnaire was administered to the head or

most knowledgeable person of the household to identify eligible respondents (men aged 18-49

years) in the sampled household. If more than one eligible man was identified in the sampled

household, only one man was selected using the KISH table. If a selected household had no

man who could be identified as eligible, sampling continued using the same approach until the

desired sample size of 25 had been reached in each cluster.

The team interviewed 1,000 men aged 18-49 years (400 urban and 600 rural). The primary

sampling unit (PSU) was a village ward or combination of wards of a village development

committee (VDC) in a rural area and a municipal sub-ward in an urban area, with at least 100

households. As targeted, 1,000 men aged 18-49 years from 1,283 households were successfully

interviewed. A district-wise sample performance is shown in annexure table 2.1.

2.3 Study tools

A structured interviewer-administered paper questionnaire was developed based on the

IMAGES survey tool and Partners for Prevention (P4P), translated into the local languages

(Nepali and Vietnamese) and pre-tested locally in both countries. Based on the pretesting

results, the questionnaires were revised and finalized. Most of the questions were close-ended

and some key variables were included in the multiple questions. The individual questionnaire

was divided into 11 sections with approximately 250 items and took about an hour to apply

in the Nepal study and around 45-60 minutes in Vietnam.

This study focuses on men’s attitudes towards gender equity, son preference, and violence

(intimate partner violence and other forms), and their knowledge about reproductive health

and existing policies. As a consequence, the survey instrument was designed to be relevant

to these issues as they applied to adult men in stable, co-habitating relationships as well as

those not in a stable relationship. Specific topics in the questionnaire were:

!

Socio-demographic characteristics:

Current age, education, marital status, dowry

practices, caste/ethnicity, religion, type of family, sources of income, employment

experience, unemployment and underemployment, stress and reactions associated with

unemployment

!

Childhood experiences

: Victimization by violence as children, childhood trauma, witnessing

of gender-based violence, gender-related attitudes perceived in family of origin, gender

patterns of childhood friendships

!

Attitudes about relationship and son preference:

Attitudes toward gender equality,

masculinity, son preference and women’s reproductive rights

!

Intimate relationship:

Decision making on households’ matter, use of violence (physical,

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

14

sexual, psychological) against partner, men’s use of sexual violence against non-

partners

!

Reproductive history of partner:

Date of last pregnancy, practices of ultrasound and

outcomes of last pregnancy

!

Preference on family size and composition:

Sex preference, importance of having son or

daughter and reasons

!

Knowledge about abortion law:

Knowledge about legal conditions for abortion, place for

safe abortion services and ever-experience of abortion

!

Fatherhood:

Number of living children and their sex, use of paternity leave, importance

of education, marriage, security, employment for sons and daughters, experiences of

humiliation for not having a son, any measures taken to have a son

!

Health and wellbeing:

Mental health issues (depression, suicidal ideation), life style-

related questions (substance or alcohol abuse)

!

Policies:

Attitudes toward various gender equality policies of the country

!

Other final questions:

Sexual experience and satisfaction with sexual life, sexual behavior,

experience of any signs and symptoms of STIs and use of/victimization of violence in

other contexts.

In Vietnam, the questionnaire included some specific questions for the Vietnamese context,

such as those related to the bride-price and the number of doctors visited for an ultrasound

test. The study excluded items related to dowry and number of wives the respondent had

simultaneously, as these were considered irrelevant to the Vietnamese context.

2.4 Conducting the survey

Pre-Survey preparations

: In both countries, the field teams underwent five-day intensive

training in the application of the research instrument and quality assurance from the core

team members, based on standard World Health Organization practices for carrying out

research on sensitive issues. There were a total of 12 interviewers in both country field teams.

Since the survey respondents were men, all the data collectors were male, because it has been

found that male interviewers are likely to get more accurate information on sensitive issues

from male respondents.

Prior to the fieldwork in the Vietnam study, the local authorities and the Department of

Population and Family Planning of the two selected provinces, Hung Yen and Can Tho, were

contacted to obtain their permission and assistance for the research team. The survey was

conducted in Hung Yen from 3-10 August, 2011 and in Can Tho from 17-26 August, 2011. In

total, 1,424 men were interviewed, 719 in Hung Yen and 705 in Can Tho.

The questionnaire was administered by the interviewers for the first 10 sections and self-

administered for the last section with more sensitive questions. Standard procedures were

followed to ensure anonymity and confidentiality.

In Nepal, the study was conducted between July and August 2011. Since CREHPA had their

own team for the survey, they did not need support in terms of manpower, but prior to the

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 15

survey, they took permission from the local authorities. The interviews were conducted at

locations convenient for the respondents, usually in a closed room in their homes. Each face-

to-face to interview took 60-90 minutes. During the field study, core team members visited

the study sites to ensure interview quality and respondents’ privacy. None of the respondents

refused to give an interview, but in a few cases, interviewers had to spend a considerable

amount of time in explanations about the study.

2.5 Analysis and variables

The selection of variables in the data analysis was informed by previous research confirming

the associations or impact of various factors as shown in framework. This report focuses mainly

on descriptive statistics and bi-variate analyses of the associations between educational levels

and economic or work-related stress and men’s gender-related attitudes, son preference and so

on. The Pearson chi-square test was used to measure the association and relations for which

p<0.05, considered statistically significant. In addition, multivariate logistic regression was

carried out on key variables of interest in the study. Most of the variables used in the analysis

are self-explanatory, but some were constructed as described in the following paragraphs.

The GEM scale

: As a measure of men’s gender-related attitudes, this study applied the Gender-

Equitable Men (GEM) scale, originally developed by the Population Council and Promundo with

young men aged 15-24 years (Barker et al, 2011) and later adopted by the IMAGES for adult

men. In the present study, a GEM scale was created consisting of 15 attitudinal statements

in Nepal and 17 statements in Vietnam about different dimension of men’s gender attitudes

(including sexuality, violence, household tasks, homophobia and male/females roles). These 15

and 17 statements were selected out of a total of 24 statements; using factor analysis and

after assessing reliability (Cronbach Alpha scores was 0.79 for Nepal and 0.88 for Vietnam) a

composite variable was constructed. The scale results of respondents were trichotomized by

their total scores and categorized as “low equity’, “moderate equity’ and “high equity’.

Economic stress index:

This is a binary index created from responses to six statements related

to stress or depression as a result of work or income deficiencies. Responses were categorized

into ‘strongly agree’, ‘agree’, ‘disagree’, and ‘strongly disagree’, and then further clubbed into

two categories of ‘disagree’ and ‘agree’, based on which a composite variable for economic

stress was created.

A S

on-preference attitude

variable was created based on 11 of the 12 attitudinal statements

about different dimensions of son-preference after factor analysis and after assessing

reliability. A similar process, as was applied to the GEM scale, was followed to construct this

variable. Based on their scores, respondents were then categorised into ‘low’, ‘medium’ or

‘high’ levels of son preference.

The

Depression index

was created based on responses to the 16 of the 17 statements

regarding men’s feelings and experiences related to various daily life problems. The scale was

developed using factor analysis and internal consistency check (Cronbach’s alpha=0.77).

A

Relationship control index

was also created (following the process for creating the GEM

scale) based on the eight statements related to different dimensions of partner control and

categorized as ‘low’, ‘medium’ and ‘high’.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

16

The

Positive deviant

indicator was developed to represent men who witnessed/experienced

an inequitable gender attitude during childhood and have high-to-moderate gender attitudes

now. This composite variable was created by combining the effects of those who experienced

gender inequity during childhood and their GEM scale score categorization.

The

Hyper masculine

behavior index, a binary index, was created by combining those who

reported physical violence and abuse of alcohol.

2.6 Ethical considerations

The study protocol was approved by ICRW’s Institutional Review Board (IRB) and by the

Institutional Review Committee (IRC) of the Centre for Research Environment and Population

Activities (CREHPA). The study team, including the research assistants were made aware of

the sensitivity of the area being explored under the study and took several steps to minimize

discomfort to study participants. Ethical guidelines developed by the WHO on studying sensitive

issues were maintained throughout the study and beyond. Participants in the study were fully

informed about the nature of the study, the research objectives, and the confidentiality of the

data, and gave written consent (a thumb-print for those who could not sign their names) for

their participation in the study. Standard procedures were followed for ensuring anonymity

and confidentiality of the participants.

2.7 Challenges and limitations of data

In Vietnam, the study was carried out in only two provinces, one in the north and one in the

south, so the data are not representative for the country as a whole. The analysis, however,

does indicate some significant patterns and trends with respect to son preference and

gender-based violence that deserve further policy attention. It should be noted here that, in

two out of four rural communes, which were randomly selected in Can Tho, the majority of

their populations originally migrated from the north as a result of the partition of the country

in 1954. Most of them are Catholics. The research team found that this generation has still

upheld northern accents, way of life and culture. The fact that the majority of the respondents

in the two southern communes have northern origins might produce biases to provincial

comparisons.

In Nepal, the study was carried out in the three districts which were selected using the sex

ratio of children under one years of age. The findings are meant to be representative at the

district level and not for the country. The fieldwork coincided with the peak monsoon, so there

were several barriers to physically accessing the target populations. Another challenge was

the high migration of men for employment, so it was difficult to find the selected respondents,

which resulted in replacements in the sample.

A general limitation of the data could arise from response bias, which is found unavoidable in

every interview. In this study, for some ‘sensitive’ questions, especially about violence and sex

selection practices, respondents may tend to give responses that they think are socially and

politically acceptable as “right” or to please the interviewer.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 17

3. SOCIO-DEMOGRAPHIC PROFILE

OF RESPONDENTS

This chapter lists the demographic and socio-economic characteristics of the population under

study in the two countries. Furthermore, there is an analysis of economic stress, partner’s

characteristics and substance abuse if any. Indicators or indices of the levels of depression,

masculinity, and relationship control were created and are discussed in this chapter. Some of

the socio-economic characteristics of their partners, if any, such as age, and education and

income levels are analyzed as these could have a bearing on the study.

3.1 Demographic characteristics

Age:

Over a quarter of the respondents in Nepal (29 percent) were below 24 years in age, and

over one-third of them (39 percent) interviewed were between 35 and 49 years (table 3.1).

In Vietnam, a smaller proportion of the men interviewed were between 18 and 24 years (19

percent) and the majority (54 percent) was in the oldest category of 35 and 49 years. The

mean age of the respondents in Nepal was 32 years versus 35 years in Vietnam.

Marital Status and Place of Residence:

The large majority of the respondents in both countries

(79 percent in Nepal and 76 percent in Vietnam) were currently married (table 3.1). Similarly,

the majority of the sampled men in both countries lived in the rural areas.

While a slight majority of the Nepali respondents (54 percent) lived in a nuclear family structure,

the vast majority of the men interviewed in Vietnam (91 percent) lived in a joint family.

Most of the Nepali men had married at a younger age as their mean age at marriage was

20 years, compared to 25 years for the Vietnamese. In fact, almost half the sample of Nepali

men (48 percent) had married before they were 20 years old; in comparison, almost half the

Vietnamese men (45 percent) had married after the age of 25.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

18

3.2 Socio-economic characteristics

Table 3.2 presents selected social characteristics of the study population. As the table shows,

the majority of men in the sample from both countries (55 percent) had studied up to the

school leaving certificate level. In Vietnam one-third of them had gone beyond the higher

secondary level, and in Nepal, this proportion was 22 percent. In the Nepali sample, 8 percent

of the men were illiterate.

The caste/ethnicity analysis of the sample of Nepali men shows that the largest proportion of

men in the sample belong to the Brahman/Chhetri group (35 percent) followed by the Janajatis

(31 percent), and the

dalits

(22 percent).

SELECTED DEMOGRAPHIC CHARACTERISTICS

TABLE 3.1

Characteristics

Nepal Vietnam

Percentage n Percentage n

Age

18-24 years 29.1 292 19.1 270

25-34 years 31.7 317 26.9 382

35-49 years 39.1 391 54.0 768

Mean age 31.7 1000 34.9 1425

Marital status

Currently not married 21.6 216 24.3 1079

Currently Married 78.4 784 75.7 346

Place of residence

Rural 60.0 600 68.5 976

Urban 40.0 400 31.5 449

Type of family

Nuclear 59.8 598 8.6 123

Joint/Extended 40.2 402 91.4 1302

Age at marriage

<20 years 48.3 383 5.7 62

21-24 years 39.7 315 49.1 434

25-49 years 12.0 95 45.2 491

Mean age 20.3 793 24.6 1087

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 19

Almost all the men in Nepal stated that they were Hindus, while in Vietnam the largest group

(64 percent) said they were non-religious. Among those who reported belonging to a religion,

16 percent were Christian, 9 percent were Buddhist, and 0.1 percent was Muslim.

Table 3.3 presents the employment and economic characteristics of the interviewed men.

Farming was the most common occupation in both countries, 54 percent in Nepal and 36

percent in Vietnam. In Vietnam the second-largest group was manual laborers followed equally

by the other two occupational groups, business/shopkeepers and professionals. In Nepal the

occupational distribution of the remaining men was a bit different, with professionals forming

the second-largest group (19 percent), followed by businessmen (15 percent) and manual

laborers (12 percent).

The overwhelming majority of the working men in both countries (80 percent in Nepal and 71

percent in Vietnam) said they work throughout the year, while less than one-fifth in Nepal (17

percent) and one-third (27 percent) in Vietnam said they work seasonally.

SELECTED SOCIAL CHARACTERISTICS

TABLE 3.2

Characteristics

Nepal Vietnam

Percentage n Percentage n

Level of education

Illiterate 8.0 80 - -

NFE/up to primary 15.1 151 11.0 157

Primary to SLC 55.2 552 55.6 792

Higher secondary and above 21.7 217 33.4 476

Caste/Ethnicity

Brahman/Chhetri 35.4 354 - -

Janajatis 30.8 308 - -

Disadvantaged non-dalit terai

caste group 12.1 121 - -

Dalits/religious minorities 21.7 217 - -

Religion

No religion - - 64.4 918

Hindu 92.1 921 - -

Christian 2.2 22 15.8 225

Buddhist 2.9 29 8.5 121

Muslim 2.8 28 0.1 1

Others - - 11.2 160

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

20

SELECTED ECONOMIC CHARACTERISTICS

TABLE 3.3

Regarding the main income earner in the family, the countries showed very different patterns.

While the majority of the interviewed men in both countries said they were the main earners,

the share varied from 65 percent in Nepal to 42 percent in Vietnam. In Vietnam, 30 percent of

the men said that both they and their partners were the main earners in the family, while this

proportion in Nepal was only 2 percent. In contrast, in Nepal, the men’s parents were cited as

the main earners in 27 percent of the cases, while this was so in only one-fifth of the cases

in Vietnam.

Dowry and Bride Price:

The survey assessed dowry practice of the men at the time of their

marriage. Of the sampled men in Nepal, 53 percent said they had not asked for a dowry or

paid a bride price during their marriage, while 45 percent reported having received a dowry.

Among those who responded that dowry was involved, 85 percent reported that it was given

as a gift and only 15 percent stated that it was demanded.

In Vietnam, of the sampled men, 90 percent said they had paid a bride price during their

marriage; of these, 97 percent reported having given this as a gift, while 3 percent said it had

been demanded by the bride’s family.

Characteristics

Nepal Vietnam

Percentage n Percentage n

Occupation

Unemployed/student - - - -

Professional/service 19.2 170 19.8 252

Manual labor 12.1 107 24.4 311

Business/shop 15.0 133 19.9 254

Farmer 53.7 475 35.9 458

Time of work

Throughout the year 80.0 708 70.9 905

Seasonally 17.1 151 27.4 350

Once in a while 1.6 14 1.4 18

Currently not working 1.4 12 0.2 3

Main earner of family

Self 64.9 649 42.9 611

Wife/partner 0.8 8 4.8 68

Both equally 2.1 21 30.1 429

Parents 27.4 274 19.4 276

Siblings 3.8 38 - -

Others (uncle/aunt /

grandparents/son/ daughter) 1.0 10 2.8 40

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 21

3.3 Economic stress

To further explore characteristics that may affect men’s attitudes towards son preference,

in addition to asking about their employment status, we also asked the currently employed

men specific questions about their work-related stress or depression. Furthermore, given the

prevailing, socially expected role of men as breadwinners of the family, economic stress may

be an indicator of men’s life experiences and a contributing factor to their attitudes toward

son preference and use of violence.

As described in Chapter 2, economic stress is an index created from six attitudinal statements

on the status of employment (table 3.4). As we can see, the majority of the sampled men in

Nepal (63 percent) stated they were experiencing economic stress, compared to 38.2 percent

of the Vietnamese men (figure 3.1).

MEN’S REPORT OF ECONOMIC STRESS

FIGURE 3.1

A fairly large proportion of men in both countries believed they were stressed or depressed

from lack of work or income (table 3.4). In almost all the cases, the proportion of Nepali men

admitting to feeling stressed or depressed was higher than the Vietnamese men. For instance,

while the largest share of men in both countries admitted to feeling stressed because they did

not have enough income, in Nepal the proportion was 58 percent, compared to 36 percent in

Vietnam. Half the Nepali men stated they were ashamed to face their family if they were out

of work, while 30 percent of Vietnamese men admitted to this. Similarly, almost half the Nepali

men said they were frequently stressed or depressed because they did not have enough work,

and 30 percent of the Vietnamese men said the same. Another important cause of stress or

depression among the respondents was the fact that they spent most of the time out of work

or looking for work (45 percent for the Nepali men and 35 percent for the Vietnamese).

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

22

Only a very small proportion of men in either country agreed with extreme positions arising

from economic stress, i.e., they had considered leaving their family because they were out of

work, or that they drink or stay away from home when they can’t find work.

3.4 Partners characteristics

In Nepal, over two-fifths of the interviewed men’s partners were between 25 and 34 years

while, a quarter (24 percent), were younger than 24 years (annex table A3.1). In Vietnam, the

majority of the partners (57 percent) were between 35 and 49 years of age, and only 8 percent

were below 24 years.

Analysis of the partners characteristics showed educational differences between the two

countries. For example, while the majority of Nepali men (61 percent) were more educated

than their partners, the largest segment in the sample of Vietnamese men (46.5 percent) said

they had the same level of education as their partners. A little over one-tenth (12 percent) of

Nepali men said their partners were more educated, while over one-fourth (26 percent) of the

Vietnamese respondents said this was the case.

A comparison of the respondents incomes with their partners showed that in 41 percent of

the cases Vietnamese men had the same income levels as their partners, while 20 percent of

Nepali men earned the same income as their partners. The overwhelming majority of Nepali

men (75 percent) earned more than their partners compared to 53 percent of the Vietnamese

respondents. A small proportion of the respondents in both countries (2 percent in Nepal and

7 percent in Vietnam) stated that their partner earned more than they did.

3.5 Alcohol and substance abuse

The majority of men in both countries said they had drunk alcohol at some time, with the

STATEMENTS OF ECONOMIC STRESS

TABLE 3.4

Statements Percentage (Yes)

Nepal Vietnam

I am frequently stressed or depressed because of not

having enough work 47.0 (n=873) 29.9 (n=1272)

I am frequently stressed or depressed because of not

having enough income 58.4 (n=873) 35.8 (1272)

I sometimes feel ashamed to face my family because I am

out of work 50.4 (n=127) 30.4 (n=148)

I spend most of the time out of work or looking for work 45.7 (n=127) 35.1 (n=148)

I have considered leaving my family because I was out of

work 3.1 (n=127) 4.1 (n=148)

I sometime drink or stay away from home when I can’t find

work 0.8 (n=127) 4.7 (n=148)

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 23

proportion ranging from 85 percent of the respondents in Vietnam to 52 percent in Nepal

(annex table A3.2). One in six of the respondents from Nepal (17 percent) and 1 percent of the

respondents from Vietnam reported having used drugs in the past year.

3.6 Other indicators

Four other indicators were created from the information obtained from the respondents as

part of the study, and the results from these are given below.

A depression scale index

was created based on 16 statements regarding the respondents

feelings and experiences on various issues of daily life in the previous week. The results

showed that slightly less than a quarter of the men in both countries (25 percent in Nepal

and 24 percent in Vietnam) reported moderate depression in the past week. The proportion of

men in the high depression category was fairly low at about 4 percent in Nepal and 1 percent

in Vietnam (figures 3.2 and 3.3).

OTHER INDICATORS - NEPAL

FIGURE 3.2

The positive deviant indicator

was developed to show those who had experienced inequitable

gender attitudes during their childhoods and now have high-to-moderate gender equitable

attitudes. As described in chapter 2, this composite variable was created by combining those

who experienced gender inequity in childhood with their GEM scale scores. The results showed

that 44 percent of the men in Nepal and 61 percent of the men in Vietnam were positively

deviated on this indicator, so they had experienced inequitable gender attitudes in childhood,

but displayed more gender equitable attitudes now. This was followed by the predicatively

deviant (41 percent in Nepal and 32 percent in Vietnam); the respondents who ‘stick to the

norms’, i.e., men who had not experienced gender inequity in childhood and had low gender

equity were the smallest group, 15 percent in Nepal and 7 percent in Vietnam.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

24

The hyper-muscularity indicator

was created by combining the responses from men who

reported perpetuating physical violence and intake of alcohol. It was found that a little over a

quarter of men (24%) had hyper-masculine behavior in Nepal and slightly more than a quarter

(26%) in Vietnam (figures 3.2 and 3.3).

A relationship control index

was created from statements related to the men’s control over

women in decision making. Most of the men in both countries, about three-quarters were

found to have a moderate relationship controlling attitude towards their partners; around

11 percent of the men in Nepal and 2.6 percent of those in Vietnam had high relationship-

controlling attitudes (figures 3.2 and 3.3).

OTHER INDICATORS - VIETNAM

FIGURE 3.3

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 25

Gender equality means that women and men and girls and boys, enjoy the same rights,

resources, opportunities and protections. It implies that women and girls have the same rights

as men and boys and they are equally entitled to participate fully in the political, social and

economic development of their communities. It does not mean that men and women are

the same, but that they have equal value and should be accorded equal privileges. There is

evidence that expanding a woman’s opportunities reduces gender inequality and accelerates

development (Dulfo E., 2005). Despite important advances towards equality, socioeconomic

differences between men and women still persist both in the developing and developed

world.

Recently, policy makers and social scientists have begun to emphasize the role and responsibility

of men and boys in promoting gender equality. This emphasis arises from the recognition that

women and men must work together in order to achieve more equitable relationships and

societies. An important step in this process is understanding men’s attitude towards gender

equality.

This chapter looks at gender-related attitudes of men in the study areas in Vietnam and Nepal,

men’s childhood experience of gender inequality and their participation in domestic duties and

maternal health care. In addition, we examine associations between these variables and a

variety of socio-demographic and other factors.

4.1 Men’s attitude towards gender equality - The Gender-Equitable Men

(GEM) Scale

An important prerequisite for achieving gender equality is changing men’s attitude towards

gender norms that they internalize and that influence their behaviors. Men’s attitude towards

gender norms were assessed using an adapted version of the Gender Equitable Men (GEM)

scale developed initially by the Horizons Program and Instituto Promundo in Brazil (Pulerwitz

J. and Barker G., 2008). This scale has been widely adapted to many countries and has shown

high levels of validity.

For this survey, 24 statements were used to measure men’s attitude towards gender equality

which were sub-categorized into: gender roles; sexuality/sexual relationships; violence;

reproductive health; masculinity and value of sons and daughters (annex table 4.1). In Nepal

and Vietnam 15 and 17 statements respectively were finally selected using factor analysis

and after assessing reliability test (Cronbach Alpha score was 0.79 and 0.81 in Nepal and

Vietnam respectively) as mentioned in chapter 2. Responses were collated under two broad

categories: agree (which combined the responses ‘strongly agree’ and ‘agree’) and ‘disagree’

(which combined the responses ‘disagree’ and ‘strongly disagree’).

4. MEN’S ATTITUDE TOWARDS

GENDER EQUALITY

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

26

Overall, the results show that men in both countries were in favor of conservative gender

roles (table 4.1). For instance, in Nepal nearly half (48 percent) agreed that ‘a woman’s most

important role is to take care of her house and cook for the family’. Likewise, an overwhelming

majority, more than four out of five men (84 percent) agreed that ‘a woman should obey her

husband’. Although only a little more than one-fifth of the men (22 percent) believed that a

‘woman’s role is to produce a son for her husband’s family’, almost all (99 percent) agreed that

‘once a woman gets married, she belongs to her husband’s family’. Interestingly, less than half

the men (43 percent) felt that ‘a man should have the final say in all family matters’.

These attitudes were even more pronounced among the Vietnamese men surveyed. A vast

majority (81 percent) felt that ‘men should have the final say in all family matters’ and almost

as many (78 percent) felt that ‘a woman’s most important role is to take care of her home

and cook for her family’. Also, almost two-thirds (62 percent) agreed that “once a woman gets

married, she belongs to her husband’s family”.

Men’s attitude towards sexuality and sexual relationships showed some variation (table 4.1).

Of the sampled men in Nepal, more than half (52 percent) agreed that ‘a woman cannot refuse

to have sex with her husband’, with nearly three-fifth (58 percent) believing that if ‘a woman

doesn’t physically fight back, it’s not rape’. Similarly, 79 percent of the men disagreed with the

view that a woman who is raped is to blame. Interestingly, in Vietnam, among the surveyed

men, almost two-thirds, (62 percent) believed that a woman can refuse to have sex with her

husband, while a little over half (55 percent) felt that if ‘a woman doesn’t physically fight back,

it’s not rape’.

Interestingly, in both countries there is less agreement on the statement that a woman’s role is

to produce a son for family compared to some of the other statements of gender attitudes.

On violence against women, while less than half (44 percent) the Nepali men surveyed felt

that ‘a woman deserves to be beaten at times’, more than three-quarters (77 percent) agreed

that if a woman does something wrong her husband has the right to punish her. These

proportions were far lower among the Vietnamese men surveyed, as only 26 percent agreed

with the first statement, and 31 percent agreed with the second. Almost half the Nepali men

believed that ‘a woman should tolerate violence in order to keep her family together’, while

only 23 percent of the Vietnamese men agreed with this. Overall, there was less agreement

with the given reasons for violence against women in Vietnam than in Nepal.

In terms of reproductive health, a majority of men (64 percent in Nepal and 74 percent in

Vietnam) felt that it was not the woman’s responsibility to avoid getting pregnant.

Norms around masculinity are very strong in Vietnam where 90 percent of men agree that

to be a man you have to be tough. In Nepal too the norms are strong with about 70 percent

men agreeing with this view.

The study results also suggest that Nepalese men have strong masculine attitudes. For

example, over two-thirds of men (67 percent) agreed with the statement that ‘if someone

insults me, I will defend my reputation with force if I have to’ and ‘to be a man, you need to

be tough’ (70 percent). Likewise, the half of the men (56 percent) said they would be ashamed

to have a homosexual son. In Vietnam the results were more mixed on these indicators: while

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 27

GEM SCALE ATTITUDINAL STATEMENTS ADAPTED FOR NEPAL AND VIETNAM

TABLE 4.1

Statements

Nepal

(n=1000)

Vietnam

(n=1424)

Agree*Agree*

Gender Roles

A woman's most important role is to take care of her home and cook

for her family 48.2 78.1

A woman's most important role is to produce a son for her

husband's family 21.6 29.8

I think that a man should have the final say in all family matters 43.0 80.8

Once a woman gets married, she belongs to her husband's family 61.7

Sexuality/ Sexual Relationship

Men need sex more than women do 45.4 52.8

A woman cannot refuse to have sex with her husband 52.1 36.6

When a woman is raped, she is usually to blame for putting herself

in that situation 20.6

If a woman doesn't physically fight back, it's not rape 58.0 55.3

Violence

There are times when a woman deserves to be beaten 43.6 26.5

If a wife/partner does something wrong her husband has the right

to punish her 77.3 30.5

A woman should tolerate violence in order to keep her family

together 50.8 23.2

Reproductive Health

It is a woman's responsibility to avoid getting pregnant 36.2 26.5

I would be outraged if my wife/partner asked me to use a condom 9.5

Masculinity

If someone insults me, I will defend my reputation, with force if I

have to 33.7

To be a man, you need to be tough 69.5 90.0

It would be shameful to have a homosexual son 55.6 29.3

Values of Sons and Daughters

A man with only daughters is unfortunate 8.1 9.7

Not having a son reflects bad karma and lack of moral virtue 9.5 10.3

Only a son can provide ancestor worship 40.8

Note: * This is out of 100 and remaining disagreed to statements.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

28

GENDER ATTITUDE OF MEN IN NEPAL (n=1000) GENDER ATTITUDE OF MEN IN VIETNAM

(n=1425)

FIGURE 4.1A FIGURE 4.1B

the overwhelming majority of men (90 percent) felt that to be a man you have to be tough,

an encouragingly large proportion (71 percent) did not think it would be shameful to have a

homosexual son and two-thirds did not feel the necessity to defend their reputation with

force if they were insulted.

On the value of sons versus daughters, men’s attitudes were similar in both countries. In

Nepal and Vietnam, an overwhelming majority did not agree with the statements that ‘a man

with only daughters is unfortunate’ (92 percent in Nepal and 90 percent in Vietnam), and that

‘not having a son reflects bad karma and lack of moral virtue’ (around 90 percent in both

countries).

In order to make the results easier to interpret, the respondents were classified into three

categories based on their GEM scale scores. These categories were “low’ (18-39), “moderate’

(40-52) and “high’ (53-72) support for equitable gender norms. (figure 4.1a and 4.1b). The

results show that in both countries the vast majority of men surveyed (77 percent in Vietnam

and 71 percent in Nepal) had moderately equitable attitudes toward the gender norms in the

GEM scale. Only 14 percent of Nepali men and 16 percent of the Vietnamese men had highly

inequitable gender attitudes. While the proportion of men in both countries with ‘low’ and

‘moderately’ equitable gender attitudes was almost the same at 86 percent in Vietnam and 84

percent in Nepal, in Vietnam, a smaller proportion of men belonged to the ‘low’ equity group

(only 7 percent against 15 percent in Nepal).

4.2 Factors associated with men’s attitude towards gender equality

The GEM scale scores were tested for their association with socio-demographic characteristics,

such as province, residence, age, marital status and education of the respondents, in bi-variate

analysis using chi-square test.

In Nepal, an analysis of the men showed that there was a clear significant relationship between

their GEM scale scores and education, as 30 percent of the men with secondary schooling

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 29

and above scored in the high equity category compared to only 12 percent with primary-

senior school education and five percent with only primary school (annex table A4.2). Illiterate

men made up the greatest proportion in the low equity category (46 percent). Occupation was

found to have an association as 28 percent of the professional men scored in the high equity

category versus only around 9 percent of the semi-skilled laborers and farmers but overall

it was not significant. Caste/ethnicity had significant association with gender equity, it was

found that except for the

Brahman/Chhetri

castes, all other castes had about one in five men

with the lowest gender equitable attitude; as a corollary, around one-fifth of the Brahmans

and chhetri’s had high GEM scale scores against only one-tenth of the janajatis and dalits/

religious minorities. (annex table A4.2).

In Vietnam, among those in the ‘high’ GEM scale score category, a larger proportion (17 percent)

was in the younger age group (18-24 years) than in the older age range (13 percent in the

35-49 age group) (annex table A4.3). Religious affiliation was found to have some relationship

with gender-equitable attitudes, as a relatively smaller share of Buddhist men scored in the

high equity category (7 percent) compared to Christian men (17 percent) and those from ‘other

regions’ (50 percent), but it was not significant. Similarly, a slightly larger share of professional

men and those in service (around 17 percent) were categorized as having high support for

gender equity compared to farmers and manual laborers (14 percent).

4.3. Childhood experience of gender inequality

Early childhood experiences of gender inequality have been found in many studies to shape

people’s views and behaviors in their later life (Gil-Gonzales et al. 2008; UNICEF 2007; WHO,

2005; Heise, 1998). In this study, we hypothesize that men who have experienced or witnessed

gender inequality during their childhood (i.e. having experienced or witnessed until 18 years of

age at least one of the acts listed in table 4.2) were more likely to perpetrate gender-based

violence as adults.

MEN’S EXPERIENCE OF GENDER INEQUALITY DURING CHILDHOOD

TABLE 4.2

Statements Nepal (n=1000) Vietnam (n=1425)

Before I reached 18: Yes No Yes No

I saw my sisters/female cousins getting less freedom

than myself and my brothers 42.3 57.7 41.1 58.9

I was told that daughters were a liability to the family 16.5 83.5 17.6 82.4

I saw the hardship my parents/relatives went through to

pay dowry 19.6 80.4 39.9 60.1

I saw my parents demanding dowry on my brother’s

marriage 14.1 85.9 38.2 61.8

I saw my sister/female cousin being ill-treated or

abandoned on being unable to produce a son 4.7 95.3 5.6 94.4

Men in Nepal and Vietnam are brought up in a socio-familial context where gender

discrimination against women is common. In both countries, the most common form of

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

30

gender inequality that men witnessed during their childhood was their sisters or female

cousins having less freedom than male counterparts in the family (42 percent in Nepal and 41

percent in Vietnam) reflecting the overall gender restrictions in society; the next most common

form in both countries was the hardship that their family or relatives went through to pay

dowry, although the reported proportion was far lower in Nepal (20 percent) versus Vietnam

(40 percent). Encouragingly, a relatively low proportion of men in both countries (5 percent in

Nepal and 6 percent in Vietnam) reported having ever seen their sister or female cousin being

ill-treated or abandoned because they were not able to produce a son.

MEN’S EXPEREINCE OF GENDER INEQUALITY

DURING CHILDHOOD IN NEPAL (n=1000)

MEN’S EXPEREINCE OF GENDER INEQUALITY

DURING CHILDHOOD IN VIETNAM (n=1425)

FIGURE 4.2A FIGURE 4.2B

The above five statements (table 4.2) were combined to create a binary index to measure

gender inequality in childhood. In both countries, the majority of men reported experiencing or

witnessing some form of gender inequality in childhood, more than half in Nepal (55 percent)

and two-thirds of men in Vietnam (66 percent) (figures 4.2a and 4.2b).

4.4 Factors associated with childhood gender inequality

To explore the factors associated with men’s childhood experience of gender inequality, bi-

variate analysis using chi-square test was carried out. In Nepal it was found that caste/

ethnicity, GEM scale score category and economic stress were significantly associated with

childhood experience of gender inequality. For example, disadvantaged non-dalit terai men

were significantly more likely to report experiences of gender inequality in childhood than other

ethnic groups, as were men who faced economic stress. Likewise, men reporting low gender

inequality experience in childhood have a higher gender equitable attitude in adulthood.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 31

MEN’S REPORTED CHILDHOOD EXPERIENCE OF GENDER INEQUALITY BY SELECTED CHARACTERISTICS

TABLE 4.3

Characteristics Nepal (n=1000) Vietnam (n=1425)

Yes (%) n Yes (%) n

Age

18-24 Years 59.6 292 58.9 270

25-34 Years 52.4 317 65.4 382

35-39 Years 51.9 203 69.5 768

P-value 0.097 <0.01

Educational Status

Illiterate 60.0 80 - -

Up to Primary 55.0 151 68.2 157

Primary to Senior school 51.8 552 70.1 792

Higher Secondary or more 58.1 217 59.2 476

P-value 0.297 <0.01

Place of Residence

Rural 55.0 600 66.5 976

Urban 53.2 400 65.7 449

P-value 0.586 0.768

Caste/Ethnicity

Brahmin/Chhetri 46.9 354

Janaties 47.4 308

Non Dalit Terai 81.8 121

Dalits/Religious minorities 60.8 217

P-value <0.01 NA

Type of Family

Nuclear 52.7 598 71.5 123

Joint/Extended 56.7 402 65.7 1302

P-value 0.209 0.194

Occupation

Unemployed/Student 53.9 115 - -

Professionals 55.9 170 63.1 252

Manual Labors 57.8 306 63.0 311

Business/Shops 52.6 133 68.5 254

Farming 50.4 276 71.1 458

P-value 0.09 <0.05

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

32

Characteristics Nepal (n=1000) Vietnam (n=1425)

Yes (%) n Yes (%) n

Economic Stress

No 47.0 321 68.8 786

Yes 59.1 552 64.6 486

P-value <0.01 0.119

GEM Scale Score Category

Low 68.7 150 77.9 140

Moderate 51.8 706 65.5 173

High 51.4 144 62.3 212

P-value <0.01 <0.01

Total 54.3 1000 66.2 1425

Apart from the GEM scale score category which was significant in both countries, in Vietnam,

a completely different set of factors were significantly associated with gender inequality

experiences. For example, age, educational status and occupation played a significant role, as

older men, less educated men, and those who worked in farming or business were more likely

than others to report experiences of gender inequality in childhood (table 4.3).

Some factors such living in a rural versus an urban area, or within joint family vs a nuclear

family appeared not to be related to gender inequality experiences in childhood.

4.5 Participation in maternal health care and child care

An indicator of men’s support for gender equality is their participation in maternal health and

child care. Men’s participation in maternal health care, as measured by their participation in

prenatal visits with their wife, was low in Nepal and a bit higher in Vietnam. Only about 40

percent of the surveyed men in Nepal reported having accompanied their wife/partner during

a prenatal visit for the most recent pregnancy, while 56 percent of the Vietnamese men did

so. While a majority of men in both countries took part in the daily care of their children, fewer

men did so in Nepal (55 percent) than in Vietnam (71 percent) (table 4.4).

Paternity leave, paid or unpaid, has been promoted as a way to increase men’s involvement

in child care. A very low proportion of employed men in Nepal took paternity leave at the

birth of their last child, and in fact, almost 60 percent of the employed men (i.e. 29 percent of

49 percent) took no leave during the birth of their last child. In Vietnam, 23 percent took no

leave, while among the 77 percent who took some leave, there was a variation in terms of the

duration of leave: around 30 percent were off work for less than two weeks, and 18 percent

for more than three months.

In Vietnam, while the Labor Code has specific provisions for maternity leave, it contains no

provision for paid paternity leave. Rather, all workers who become new fathers are entitled

to an unpaid leave for personal reasons. The lack of a specific legal framework for paternity

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 33

leave may perpetuate an unequal division of labor within the family and discourage men’s

participation in the care of children. A fairly high percentage of Vietnamese men taking time

off to take care of their newborn baby and their wife (77 percent), together with the relatively

high proportion accompanying wives to prenatal care (56 percent), indicates that men in the

survey care about maternal and child health.

There appears to be a clear relationship between men’s participation in maternal health care

and child care and socio-demographic characteristics (annex table A4.4). Men in both countries

who accompanied their wives/partners on their prenatal visits were more likely to be younger,

live in urban areas (as opposed to rural), have a higher level of education and be professionals

(as opposed to farmers). In Nepal, other characteristics that showed a significant relationship

with accompanying partners for prenatal visits were scoring in the high equity category on

the GEM scale, lower levels of depression and economic stress and caste/ethnicity, where

men who were from the Brahman/chhetri caste were more likely to do so than janajatis

(this was not relevant for Vietnam). In Vietnam, none of these other factors had a significant

relationship with men’s participation in prenatal visits.

Unsurprisingly, in both countries there was a significant relationship between men’s daily

participation in the care of children and their GEM scale scores. In addition, in Nepal, a

significantly larger proportion of men in nuclear families and who worked in farming helped

with the daily care of children, while in Vietnam, men’s level of education had a significant

relationship with their involvement in the daily care of children.

PARTICIPATION IN MATERNAL HEALTH CARE AND CHILD CARE

TABLE 4.4

Participation Nepal

(n=714)

Vietnam

(n=999)

Whether or not accompanied to wife/partner during prenatal visit for the

most recent pregnancy

Yes 39.5 56.4

Whether participated in daily care of a child

Yes 55.1 71.0

Whether took leave/time off during the birth of the last child

No leave 28.9 20.4

Less than one week 7.3 10.3

1-2 weeks 8.0 19.2

3-4 weeks 2.0 9.1

1-3 months 2.7 11.7

More than 3 months 0.3 18.4

Not employed/not applicable 51.0 11.0

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

34

PARTICIPATION IN DOMESTIC DUTIES

TABLE 4.5

4.6 Men’s participation in domestic duties

Men’s participation in household duties was fairly low in both countries, as domestic work is

usually borne by their wives in a majority of the cases (63-67 percent). Over one-tenth of the

men in Nepal (11-13 percent) reported that their partners did all the work, while in Vietnam,

the percentage of men whose wives did all the work was 4 to 10 percent.

Interestingly, around one-fifth (20-25 percent) of the men in both countries reported sharing

the work equally or doing the work together with their partners (table 4.5).

Participation of

respondent

Type of domestic work

Preparing food Cleaning the house Washing clothes

Nepal Vietnam Nepal Vietnam Nepal Vietnam

I do everything 0.3 0.9 0.3 0.6 0.4 0.6

Usually me 0.1 6.6 0.8 5.4 0.5 3.5

Shared equally or

done together 19.3 23.3 21.4 26.1 21.2 21.2

Usually partner 66.6 64.1 65.5 63.1 64.9 64.6

Partner does

everything 13.0 4.2 11.3 4.0 12.3 9.4

Neither of us do/

we do not live

together

0.8 0.6 0.8 0.8 0.8 0.8

n 794 1101 794 1101 794 1101

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 35

Violence against women is one of the most extreme manifestations of gender power inequalities

and is used by many men to exert control and dominance over women. This study assessed

men’s perpetration of different types of violence against their intimate female partners in

the two countries under study. In addition, the determinants of violence against an intimate

partner are identified and discussed.

5.1 Types of violence reported against a wife/partner

The sample of men was asked a series of questions to assess the prevalence of violence

perpetrated by them against their intimate partners. The questions covered acts of emotional

violence (five questions), economic violence (three questions), physical violence (five questions)

and sexual violence (four questions).1 The responses to each item were combined to create a

composite index for each type of violence.

All men who currently have or ever had a wife/female partner were asked whether they had

committed different types of violence against her, including physical, emotional, economic and

sexual violence. The term ‘violence’ did not appear in the survey, instead, a series of questions

was asked to measure different acts of violence. The specific questions for the four areas of

violence are given below:

Physical violence to wife/partner

(1) Slapped a wife/partner or threw something at her that could hurt her

(2) Pushed or shoved a wife/partner in anger

(3) Hit a wife/partner with a fist or with something else that could hurt her

(4) Kicked, dragged, beaten, choked, or burned a wife/partner

(5) Threatened to use or actually used a gun, knife, or other weapon against a wife/partner

Emotional violence to wife/partner

(1) Insulted a wife/partner or deliberately made her feel bad about herself

(2) Belittled or humiliated a wife/partner in front of other people

(3) Did things to scare or intimidate a wife/partner on purpose, for example, by the way you

looked at her, by yelling and smashing things

(4) Threatened to hurt a wife/partner

1 See table A5.1 in the annexure for the list of questions and responses by country.

5. MEN’S ATTITUDE TOWARDS AND

EXPERIENCE OF VIOLENCE

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

36

PREVALENCE OF KIND OF VIOLENCE AGAINST WIFE/PARTNER

FIGURE 5.1

Economic abuse to wife/partner

(1) Prohibited a partner from getting a job, going to work, trading or earning money

(2) Took a wife/partner’s earnings against her will

(3) Threw a wife/partner out of the house

Sexual violence to wife/partner

! Had sex with wife/partner when you knew that she did not want it but you believed she

should agree because she was your wife/partner

! Forced wife/partner to have sex with you when she did not want to

! Forced wife/partner to watch pornography when she did not want to

! Forced wife/partner to do something sexual that she did not want to

For the purpose of analysis, if the man committed at least one of the listed abusive acts

in a particular category, he was considered to perpetrate that specific form of violence. The

questionnaire looked at two periods of time when the acts were committed: during the

respondent’s lifetime and in the 12 months preceding the interview. For each of the lifetime

abusive acts, a follow up question was asked about the frequency with which the violence had

happened: once, few or many times.

Overall, the results indicated that most forms of violence against women are prevalent in both

countries. In Nepal, about 71 percent of men reported ever using any form of violence against

their intimate partner, while in Vietnam, the proportion was 60 percent.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 37

The most common form of intimate partner violence (IPV) in both countries was emotional

violence, which almost half the sampled men in both countries reported having enacted at

some time. The second most common form was physical violence - more than two-fifth of

the Nepali men (42 percent) reported being physically violent, while in Vietnam this form of

violence was reported by one-third of the sampled men. One-fourth of the Nepali men stated

that they had enacted sexual violence, while the proportion of men reporting it in Vietnam

was only around one-tenth; economic violence was lowest among Nepali men (only around

10 percent), while one-fifth of the sampled men from Vietnam reported resorting to economic

violence at some time.

The responses to violence against women in the past year showed that more than 40 percent

of men in Nepal and 25 percent of the Vietnamese men had committed some form of violence

in the past year. In both countries, the most common form of violence in the last year, as is

the case with ever use, was emotional violence, followed by physical violence.

The data from Vietnam are consistent with the results from the national survey on gender-

based violence conducted with female respondents by the General Statistical Office (GSO) in

2009 which revealed that 58 percent of ever-partnered women had experienced at least one

form of violence and 27 percent in the past 12 months (GSO, 2010).

As mentioned above, the most commonly perpetrated form was emotional violence in both

countries - in Vietnam it took the form of insulting a wife or partner so that they felt bad about

themselves (38 percent), while in Nepal, most men had been abusive to scare or intimidate

their wife/partner (33 percent) (annex table A5.1). Other common violent acts were pushing

or shoving a wife/partner in anger (31 percent in Nepal), slapping a wife/partner or throwing

something that could hurt her (28 percent in Vietnam and 24 percent in Nepal) and having sex

with a current partner when she did not want it (24 percent in Nepal).

5.2 Violence by selected socio-demographic characteristics

Table 5.1 indicates the percentage distribution of men perpetrating lifetime IPV by select socio-

demographic characteristics. The analysis shows that with an increase in educational level the

proportion of men perpetrating IPV tends to decline in both countries. In Vietnam, age, too

has a significant impact on the use of violence, as the proportion of men reporting violence

increases with their age. Thus in Vietnam, older men with fewer years of education are more

likely to perpetrate violence against their spouse/wife. In contrast, in Nepal, age, place of

residence, and type of family appeared to have no correlation with the lifetime perpetration

of violence. Occupation had some correlation with IPV as a lower proportion of professional

men in both countries reported using violence compared to other professions such as manual

laborers and traders. In Nepal, a high proportion of unemployed men reported using IPV.

Interestingly, in Vietnam the family structure played a role in perpetration of violence against

female partners, as men in joint or extended families were far more likely to engage in violent

acts both in their lifetime (62 percent) as well as in the recent past (27 percent), than men in

nuclear families (43 percent for lifetime and 12 percent in the recent past). No such association

was evident in Nepal (table 5.1 and annex table A5.2).

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

38

Characteristics Nepal (n=1000) Vietnam (n=1425)

Ever n Ever n

Age

18-24 Years 70.2 131 41.5 106

25-34 Years 69.6 286 56.2 340

35-39 Years 72.1 380 65.0 766

P-value 0.764 <0.01

Educational Status

Illiterate 83.8 74 - -

Up to Primary 77.2 145 65.8 149

Primary to SLC 70.1 425 63.5 705

Higher Secondary or more 60.8 153 51.9 360

P-value <0.01 <0.01

Place of Residence

Rural 69.7 488 60.6 836

Urban 72.8 309 59.8 378

P-value 0.192 0.777

Type of Family

Nuclear 71.4 479 42.4 85

Joint/Extended 70.1 318 61.7 1129

P-value 0.378 <0.01

Occupation

Unemployed/Student 72.7 33 - -

Professionals 57.2 138 47.4 211

Manual Labors 78.9 252 67.5 268

Petty trade/shop 80.2 121 61.8 241

Farming 65.3 248 63.1 439

P-value <0.01 <0.01

Total 70.9 - 25.9 -

PERPETRATION OF INTIMATE PARTNER VIOLENCE BY SELECTED SOCIO-DEMOGRAPHIC

CHARACTERISTICS

TABLE 5.1

In Nepal, where data were available by ethnicity/caste, there was an association with reported

IPV highest among the indigenous (

janajatis)

men (76 percent), followed by men from the

Dalit/religious minority (72 percent) (annex table A5.2).

Data on recent violence (annex table A5.2) shows that men’s current age, level of education,

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 39

caste/ethnicity, and occupation were significantly associated with past year perpetration

of violence in Nepal. Interestingly, in contrast to the findings from Vietnam where violence

increased with men’s age, in Nepal, the proportion of men who had perpetrated violence in

the recent past decreased with their age. Responses from Vietnam on recent violence show

a correlation between men’s age, education levels, and occupation and perpetration of partner

violence in the past year.

5.3 Perpetration of violence by selected background characteristics

There is substantial research that suggests that the determinants of gender based violence

are related to men’s experience of inequities as children, their current level of economic stress

and some correlates such as substance abuse. The analysis presented in table 5.2 shows five

key predicting variables: their level of gender equity based on their GEM scale score, childhood

experiences of gender inequity, work-related stress, alcohol use, and relationship control

index2. These are all found to be significantly associated with men’s reports of violence against

intimate partners in one or both countries.

In line with expectations, a clear relationship was observed between GEM scale scores and

report of IPV in both countries (table 5.2). The higher the GEM scale score, the less likely are

respondents to report violence against their wife/partner. In other words, men who report

more equitable gender attitudes are also less violent with their partner. This was more obvious

in Nepal, where the proportion of men reporting IPV went from 84 percent (among those low

on the GEM scale) to 55 percent (among those high on the GEM scale) than in Vietnam where

the corresponding proportions of men went from 70 percent (low on GEM scale) to 55 percent

(high on GEM scale).

Unsurprisingly, the results showed that men who had experienced or witnessed gender

inequitable behavior in childhood were more likely to perpetrate IPV than those who had not

(75 percent vs. 66 percent in Nepal; and 63 percent vs. 54 percent in Vietnam).

The use of alcohol was also found to have positive significant effects on the perpetration

of intimate partner violence during men’s lifetime in both countries. While men who ever

experienced stress because of not having enough work or income were more likely to be violent

against a female partner in Nepal, there was no relationship in Vietnam. There appeared to be

no relationship between the relationship control index and the likelihood of reporting the use

of violence against a female partner in Nepal.

An analysis of the results related to recent violence (annex table A5.3) shows that the

relationships between the indicators and the likelihood of violence is almost the same as

those for ‘ever’ having perpetrated IPV: higher levels of violence were related to less gender

equitable attitudes, greater experiences of gender inequity in childhood, higher use of alcohol,

and economic stress among men. However, the proportion of men perpetrating IPV was lower

in Vietnam than in Nepal on almost all the indicators: thus, a larger proportion of Nepali men

was likely to have been recently violent with their partners had lower levels of the GEM scale

compared with Vietnamese men (47 percent vs. 34 percent); larger proportion of men in Nepal

2 Relationship control index was constructed by summing 8 variables indicating relationship control and then

trichotomizing the scores into low, moderate and high categories.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

40

PERPETRATION OF INTIMATE PARTNER VIOLENCE BY SELECTED BACKGROUND CHARACTERISTICS

TABLE 5.2

Characteristics Nepal (n=1000) Vietnam (n=1425)

Ever n Ever n

GEM scale score category

Low equity 84.3 127 69.8 126

Moderate equity 70.7 573 60.0 918

High equity 54.6 97 55.3 170

P-value <0.01 <0.05

Experience of gender inequity

in childhood

Yes 75.4 427 63.4 818

No 65.7 370 54.0 396

P-value <0.05 <0.01

Relationship control index

Low 67.4 89 64.8 145

Moderate 71.7 584 58.3 940

High 69.4 124 70.5 129

P-value 0.768 <0.05

Ever abuse of alcohol

Yes 77.3 454 61.2 1208

No 62.4 343 55.2 217

P-value <0.01 0.087

Economic Stress

Yes 74.3 479 60.9 486

No 64.5 276 61.4 786

P-value <0.01 0.465

Total 70.9 -25.9 -

who were recently violent with their partners had childhood experiences of gender equity (41

percent vs. 28 percent), had used alcohol (46 percent vs 27 percent) and faced economic stress

(44 percent vs. 29 percent).

5.4 Factors associated with lifetime violence

The relationships observed in bi-variate analysis were reassessed by using multivariate

analysis to identify important determinants, adjusting for the confounding effects of other

factors. The results of the logistic regression where the dependent variable is “perpetration of

any violence (ever)’ and the odd ratios corresponding to the associated factors are presented in

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 41

table 5.3 along with their significance levels. In this analysis, we have considered the following

determinants of lifetime IPV: age, education levels, caste/ethnicity, occupation, marital status,

GEM scale, experience of gender inequity and bullying in childhood, partners relative education

level and earnings and alcohol use.

The results (table 5.3) show most of these indicators are significant predictors of men’s ever-

use of violence against their intimate partners, some more strongly than others. Age, for

example, is a strong and significant predictor, as older men are more likely to report ever use

of IPV than younger men in both countries. This likelihood is especially strong in Vietnam,

where men aged 35-49 years and 24-35 years are more than twice as likely (2.24 times and

2.13 times, respectively) to perpetrate IPV than men aged 18-24 years; similarly in Nepal, men

aged 35-49 years are 1.73 times more likely to report ever use of violence compared to the

men in the age group 18-24 years. The likelihood of IPV among men aged 24-35 compared to

men aged 18-24 years is not significantly different. In Nepal, married men were twice as likely

to be violent towards their partners (2.01 times) while in Vietnam marriage did not appear to

affect the violence perpetrated against the partner.

Occupation levels too have strong and significant effects on IPV: businessmen and those

working in or having a shop are far more likely (2.47 times in Nepal and 1.67 times in Vietnam)

to perpetrate violence than those in a profession. The likelihood is also high among manual

laborers in Vietnam, who are more than twice as likely (2.1 times) as professionals to use

violence against their partners.

The association of violence and education was in the expected direction. The likelihood of

perpetuating violence declined with the increasing level of education in both the countries, but

the results in the model were not significant.

As expected, the GEM scale score has a strong effect on the likelihood of using violence. The

higher the GEM scale score, the less likely are the respondents to enact violence against their

wife/partner. In other words, men who report more equitable gender attitudes are also less

violent with their partner, especially in Nepal, where the odds ratio was only 0.32 times as

likely at the high end of the GEM scale and in Vietnam it was 0.59 at the high end.

Partners’ earnings and education levels have a significant bearing on the violence they face,

but interestingly the results are contradictory in the two countries. A partner with a lower

level of education than the man was more likely (1.44 times) to face violence in Vietnam, while

in Nepal she was only half as likely to (0.51 times). In contrast, a partner with lower earnings

was more likely to face violence in Nepal (1.49 times), while in Vietnam they were less likely

to face violence (0.65 times), but results were not significant in the model.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

42

ODDS RATIOS FROM LOGISTIC REGRESSION FOR DETERMINANTS OF INTIMATE PARTNER VIOLENCE

EVER

TABLE 5.3

Characteristics Odds ratio for Nepal Odds ratio for Vietnam

Age

18-24 1.00 1.00

25-34 1.47 2.13*

35-49 1.73* 2.24*

Education

Illiterate 1.00 -

Upto primary 0.66 1.00

Primary to secondary 0.57 0.82

Higher secondary and above 0.47 0.63

Caste/Ethnicity

Brahman/Chhetri 1.00 -

Janajatis 1.01 -

Disadvantaged non dalit terai caste group 0.63 -

Dalits/religious minorities 0.82 -

Occupation

Professional 1.00 1.00

Manual labor 1.63 2.10**

Business/shop 2.47** 1.67*

Farming 1.34 1.58*

Marital status

Not married 1.00 1.00

Married 2.01* 1.23

GEM scale

Low 1.00 1.00

Moderate 0.57* 0.73

High 0.32** 0.59*

Experience of gender inequality in childhood

Yes 1.00 1.00

No 0.59** 0.67**

Partner has higher education

Yes 1.00 1.00

No 0.51* 1.44*

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 43

Characteristics Odds ratio for Nepal Odds ratio for Vietnam

Partner has higher earnings

Yes 1.00 1.00

No 1.49 0.65

Bullied in childhood

Yes 1.00 1.00

No 0.51** 0.74*

Ever alcohol use

Yes 1.00 1.00

No 0.56** 0.66*

-2 Log likelihood 817.31 1360.39

Note: *: significant at 95 percent; **: significant at 99 percent

Unsurprisingly, childhood experiences of bullying and gender inequity played a strong role in

both countries, as men with experiences of either were almost twice as likely to engage in

IPV, especially in Nepal. Similarly, the use of alcohol increased the likelihood of IPV in both

countries.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

44

This chapter explores men’s attitude towards son preference and its correlates. The perceived

importance of having a daughter and a son and the different reasons for these have been

analyzed, along with socio-economic and background characteristics that are significantly

related to son preference attitudes in the two countries.

Previous studies have shown that in both Nepal and Vietnam there is a considerable level of

son preference. Among the many possible explanations for the existence of son preference in

Vietnam, the two most frequently cited reasons are that parents are typically supported in

their old age by their son(s), whereas girls usually move away from their families, and that

sons are needed to maintain the family line (Leone T, Matthews D et al., 2003). In Nepal sons

also are very highly prized because they continue the family name, can perform funeral rituals

and are expected to provide support in old age (Haughton J and Haughton D, 1995).

6.1 Son preference attitude

Son preference is the desire to have one or more sons. Measuring son preference directly is

challenging. In this study we used a four point scale to assess respondents’ agreement with

12 attitudinal statements on different aspects of son preference, including the value of sons,

devaluation of daughters, responsibility for the sex of the child, and the consequences for

sonless women.

The results (table 6.1) showed mixed attitudes towards son preference. Men had strong son

preference attitudes for socio-economic reasons but son preference was weaker on more

extreme statements such as aborting a female fetus or putting a female child up for adoption.

Not surprisingly, most men in both countries supported son preference statements related to

the direct value of having sons. The highest proportion of men agreed with the view that sons

are important to carry on the family lineage (75 percent in Nepal and 69 percent in Vietnam)

and for support in old age (60 percent in Nepal and 49 percent in Vietnam). Fewer men agreed

with the views related to aborting a female fetus (around 2 percent in either country), putting

a daughter up for adoption, or abandoning a wife who does not bear sons (3 percent in Nepal

andtwo percent in Vietnam). There was also little support for statements that denoted a lower

value for daughters, i.e. that they are a financial burden (around 10 percent in each country)

or considered a misfortune.

Interestingly, two-fifths of the Vietnamese men and a third of men in Nepal agreed with the

view that fathering a son was an indicator of their masculinity. A lower proportion in Vietnam

(13 percent) believed that it is important for a girl to marry outside her village. The proportion

was much higher in Nepal (50 percent) suggesting that exogamy was much more customary

in Nepal.

6. MEN’S ATTITUDE TOWARDS

SON PREFERENCE

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 45

PERCENTAGE OF MEN AGREEING WITH STATEMENTS ABOUT SON PREFERENCE

TABLE 6.1

A composite index on son preference attitude was developed with men’s responses to the

above statements. In both countries 11 statements were selected for the construction of the

composite index, which was done using factor analysis and reliability testing for consistency as

explained in the methodology section in chapter 2. The scores on the index were trichotomized,

reflecting high, moderate and low levels of son preference. The results showed that in both

countries, nearly three-quarters of the men (68-69 percent in both countries) showed moderate

son preference attitude with about a third of men with high son preference in both countries

(figure 6.1).

Statements Nepal

(n=1000)

Vietnam

(n=1425)

It is important to have a son to carry on the lineage or family

name

74.9 68.6

It is important to have a son to take care of you in your old

age

59.6 48.6

It is important for a woman to marry outside her village/city 50.3 13.1

The man is responsible for sex of the child 37.1 26.7

Fathering a male child shows you are a real man 31.4 40.7

The women is responsible for sex of the child 12.2 12.9

Having a daughter is financial burden/loss 9.9 10.7

A couple who has only a female child is unfortunate 8.0 9.3

If a wife partner does not have a son, her husband had good reason

to leave her or divorce her

3.1 2.1

If a wife/partner does not have a son, a family has good reason to

pressurize a man to leave his wife/partner

3.2 1.6

A couple have a good reason to abort a pregnancy if they learn it

is a girl child

1.6 2.2

A couple has a good reason to put female child for adoption 2.0 2.5

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

46

MEN’S ATTITUDE TOWARDS SON PREFERENCE

FIGURE 6.1

6.2 Factors associated with a son preference attitude

One of the primary aims of this study was to explore the factors that contribute to strong

attitudes for son preference among men. Bi-variate and multivariate analysis was conducted

using “strong’ or “moderate’ son preference attitude as the dependent variable. We examined

the effect of the respondents’ socio-demographic and other background characteristics (such

as age, years of schooling, religion, marital status and working status, relationship control

index, GEM scale score etc.) on son preference attitudes.

The bi-variate analysis of son preference by socio-economic characteristics suggests that in

both countries the level of education and type of occupation were significantly associated

with son preference attitude (table 6.2). Thus, there was a clear inverse relationship between

education and son preference attitudes, as men with higher levels of education displayed a

lower preference for sons. Similarly, in the case of occupation, manual laborers and farmers

had the highest proportion of men with high/moderate son preference attitudes. Additionally,

in Nepal, age was statistically associated with son preference attitudes among men, with older

men revealing high or moderate son preference; caste too was significant as the disadvantaged

non-dalit castes, dalit caste and religious minority castes had a higher proportion of men with

high son preference.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 47

SON PREFERENCE ATTITUDE BY SELECTED DEMOGRAPHIC AND BACKGROUND CHARACTERISTICS

TABLE 6.2

Characteristics

Nepal Vietnam

High/Moderate n High/Moderate n

Age

18-24 years 66.8 292 71.5 270

25-34 years 72.6 317 73.6 382

35-49 years 77.0 391 73.0 768

P-value, n <0.05 1000 0.83 1425

Marital status

Not married 69.4 206 73.7 1079

Married 73.4 794 72.7 346

P-value, n 0.23 1000 0.38 1425

Level of education

Illiterate 92.5 80 0.0 0

NFE/up to primary 87.4 151 80.3 157

Primary to SLC 70.7 552 76.0 792

Higher secondary and above 59.9 217 65.3 476

P-value, n <0.01 1000 <0.01 1425

Place of residence

Rural 73.0 600 72.3 976

Urban 72.0 400 74.2 449

P-value, n 0.39 1000 0.26 1425

Caste/Ethnicity

Brahman/Chhetri 62.1 354 - -

Janajatis 77.6 308 - -

Disadvantaged non dalit terai

caste group 79.3 121 - -

Dalits/religious minorities 78.8 217 - -

P-value, n <0.01 1000 -

Type of family

Nuclear 74.4 598 66.7 123

Joint/Extended 69.9 402 73.5 1302

P-value, n 0.06 1000 0.07 1425

Occupation

Unemployed/Student 66.1 115 - -

Professional/Service 65.9 170 69.8 252

Manual labor 62.6 299 80.4 311

Business/shop 72.9 133 76.0 254

Farmer 78.7 276 69.7 458

P-value, n <0.01 893 <0.01 1275

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

48

Apart from the socio-economic characteristics of the respondents, other background

characteristics revealed significant associations between men’s attitude towards son preference

and their scores on the GEM scale and the relationship control index in both countries (table

6.3). Predictably, the bi-variate chi-squaretest between GEM scale score category and son

preference attitude showed clear directionality and relationship, as men with low equitable

attitudes had a higher son preference. Likewise, the relationship control index had a significant

and positive relation with son preference attitude: thus, a larger proportion of men with high

relationship control attitudes had a high/moderate son preference attitude.

Other background characteristics that were significant only in Vietnam were experience of

gender equity in childhood, hyper-masculinity and economic stress. Men’s early childhood

experience around gender equity had a relatively strong negative association with their attitude

towards son preference, i.e., men who had experienced or witnessed gender inequity in their

childhood were more likely to prefer sons. Similarly, a higher proportion of men who reported

hyper-masculine behavior were likely to have a high son preference attitude, as were men

who experienced economic stress.

These two factors were not statistically significant among the sample of Nepali men. However,

in Nepal, son preference was significantly associated with depression levels. Thus, a higher

proportion of men with high depression or moderate depression levels were more likely to

have a high preference for sons.

The bi-variate relationships observed were further tested through multivariate analysis using

a logistic regression to identify adjusted associations between son preference attitude and

key variables such as support for gender equity and relationship control. A logistic regression

model was run, taking into account all the variables found to be significant in the bi-variate

analysis, with no strong correlation with each other. A forward stepwise regression method

with likelihood ratio criteria was used to get the best-fit model.

The results showed that son preference attitude varies significantly in both countries by men’s

relationship control, experiences of gender inequity in childhood and their support for gender

equity as measured by the GEM scale. The association was especially strong and significant in

the case of relationship control, where men with moderately high control over their partners

were three times as likely to have a son preference attitude in Nepal, and almost five times

as likely in Vietnam compared with men with low relationship control. Interestingly, men with

high control over their partners had lower son preference attitudes than those with moderate

control.

As expected, men who scored higher on the GEM scale had far lower son preference attitudes

than those who scored lower on the scale. The values were almost identical in both countries,

with men with high gender equitable attitudes 97-98 percent less likely to report son preference

attitude compared to men with low gender equitable attitudes, and 77-78 percent less likely

than men who scored moderately on the GEM scale.

Interestingly, experiences of gender inequity in childhood had contrary effects on men in

the two countries: in Nepal men who have not experienced or witnessed gender inequity in

childhood are 1.5 times more likely to have a son preference attitude, while in Vietnam, the

results are as expected.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 49

SON PREFERENCE ATTITUDE BY SELECTED BACKGROUND CHARACTERISTICS

TABLE 6.3

Characteristics

Nepal Vietnam

High/Moderate n High/Moderate n

GEM Scale

Low equity 96.0 150 95.7 140

Moderate equity 76.2 706 78.8 1073

High equity 30.6 144 27.8 212

P-value, n <0.01 1000 <0.01 1425

Experience of gender inequity

in childhood

Yes 72.7 543 75.7 944

No 72.4 457 67.4 481

P-value, n 0.48 1000 <0.01 1425

Depression scale

High 70.3 37 72.2 25

Moderate 80.6 248 76.8 342

Low 69.9 715 71.6 1058

P-value, n <0.01 1000 0.17 1425

Relationship control index

High 87.8 90 89.9 129

Moderate 77.2 626 75.1 941

Low 43.0 128 44.8 145

P-value, n <0.01 844 <0.01 1215

Hyper masculine behavior

Yes 74.5 235 82.4 102

No 72.6 609 66.2 551

P-value, n 0.26 844 <0.01 653

Economic stress

Yes 73.7 552 79.2 786

No 74.1 321 70.1 486

P-value, n 0.38 873 <0.01 1272

Total 72.6 - 72.9 -

In Nepal, age, caste and occupation also had a significant effect on men’s attitudes towards

son preference. Thus, older men aged 35-49 years were 2.5 times more likely to have son

preferring attitudes than younger men aged 18-24 years. Similarly, men who belong to the

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

50

ODDS RATIOS FROM LOGISTIC REGRESSION FOR DETERMINANTS OF ATTITUDE TOWARDS SON

PREFERENCE

TABLE 6.4

Characteristics Odds ratio for Nepal Odds ratio for Vietnam

Age

18-24 1.00 1.00

25-34 2.05* 1.00

35-49 2.45** 0.84

Education

Illiterate 1.00 -

Upto primary 0.85 1.00

Primary to secondary 0.41 0.75

Higher secondary and above 0.25 0.46*

Caste/Ethnicity

Brahman/Chhetri 1.00 -

Janajatis 1.50 -

Disadvantaged non-dalit terai caste group 3.82** -

Dalits/religious minorities 1.43 -

Occupation

Professional 1.00 1.00

Manual labor 0.32** 1.24

Business/shop 0.74 1.29

Farming 0.64 0.69

Marital status

Not married 1.00 1.00

Married 0.42 0.44*

Type of family

Nuclear 1.00 1.00

Joint/extended 1.01 1.98*

GEM scale

Low 1.00 1.00

Moderate 0.23** 0.22**

High 0.03** 0.02**

Relationship control

Low 1.00 1.00

Moderate 3.04** 4.95**

High 2.65** 2.71**

Experience of gender inequity in

childhood

Yes 1.00 1.00

No 1.51** 0.57**

-2 Log likelihood 716.43 1067.17

Note: *: significant at 95 percent; **: significant at 99 percent

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 51

disadvantaged, non-dalit, terai caste groups were almost four times (3.8 times) more likely

to have son preferring attitudes than the higher caste Brahmin/Chhetri ethnic groups.

Interestingly, professional men were most likely to have a son preference attitude than manual

laborers who were only one-third as likely to have a son preferring attitude.

In contrast, education, marital status and type of family significantly influenced son preference

among the Vietnamese men. Those educated beyond the higher secondary level were only half

as likely to have son preference attitude compared to those with only a primary education.

Men after marriage are only half as likely to prefer sons as unmarried men but results were

not significant in Nepal. Those living in a joint or extended family are two times more likely to

display son preference than men in a nuclear family.

6.3 Perceived importance of having a boy or a girl child

The patriarchal nature of Nepalese society combined with socio-economic and religious

values creates a strong influence to have a male child in the family. Similarly, in Vietnam the

patrilineal and patrilocal kinship system tends to place strong normative pressure on couples

to produce at least one son. In the survey, we explored men’s perceptions on the importance

of having a girl and a boy child.

PERCIEVED IMPORTANCE OF HAVING A BOY CHILD OR GIRL CHILD

FIGURE 6.2

Figure 6.2 displays the distribution of men by perceived importance of having a girl child and

a boy child. In Nepal, an overwhelming majority of men consider having at least a son and

at least a daughter as very important, although son preference in this context is higher than

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

52

daughter preference (83 percent vs 77 percent). In contrast, in Vietnam, the men are much

more equally distributed among the three groups – ‘very important’, ‘somewhat important’

and ‘not important’ – on the issue of having at least a son and having at least a daughter

Interestingly, in Vietnam, while almost twice as many men considered it ‘very important’ to

have a son as compared to a daughter (44 percent vs 26 percent), more men considered it

‘somewhat important’ to have a daughter than a son (46 percent vs 30 percent).

We also explored why men felt that it was important for them to have a daughter or a son?

The specific question asked about reasons for perceiving an importance in having

at least one

girl child or

at least one

boy child in the family. The results indicate that men in the survey

value sons and daughters in very different ways. This was consistent in both countries and

there were differences across the two countries.

REASONS FOR HAVING AT LEAST ONE GIRL

FIGURE 6.3

The findings demonstrated that girls were considered important for emotional support and

sharing the workload, while sons were perceived to be important mainly for socio-economic

and religious reasons (figure 6.3 and table 6.5). However, there was a difference in the two

countries in the importance ascribed to daughters – while an overwhelming share of Vietnamese

men (76 percent) cited the most important reason for having a daughter was for emotional

support, the most frequently cited reason among Nepali men was for performing religious

rituals, such as

rakshya bandan/bhai tika

(51 percent) (figure 6.3). Interestingly, in Vietnam,

girls were seen as a continued support to the family, as a large proportion of Vietnamese

men valued daughters for their emotional support and support in old age and when their

parents were sick. A smaller proportion of Nepali men valued girl children for these reasons;

this reflects Nepali traditions where girls typically leave their paternal homes after marriage

and are not expected to support their natal homes after marriage.

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 53

PERCEIVED REASONS FOR HAVING A BOY

TABLE 6.5

Men’s views about the importance of sons and daughters are highly influenced by traditional

customs and gender roles that dictate only boys can carry on their father’s name and continue

the family lineage while girls provide emotional support and are expected to be dutiful and

hardworking. This was clearly indicated by the survey results which showed that the most

important reason for having a son in both countries was the need to carry on the family

name, although the proportion of Vietnamese men citing this reason was overwhelming larger

(at 71 percent) than in Nepal (58 percent). The second and third most important reasons were

support in old age, followed by the performance of religious rituals (funeral rites for Nepali

men and ancestor worship for the Vietnamese). Sons were also valued in both countries for

the care they provided when sick and to share the workload burden. While social status and

protecting the family property were also listed as reasons to value sons among the Nepali

men, these were not important among the Vietnamese.

Reasons to have at least one boy Having at least one boy

Nepal Vietnam

Lineage 58.0 70.7

Old age support 50.7 51.3

Funeral rites 30.4 16.9

Care when parents are sick 27.4 29.7

Social status 25.2 5.6

Protecting family property 22.2 7.3

Sharing workload 20.7 22.4

Bring prosperity into family 15.2 5.2

Emotional support 11.2 28.1

Ancestor worship 11.5 48.5

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

54

Reproductive health research, programs and policies have focused almost exclusively on women

over the years. Recently, however, men’s roles and involvement in women’s reproductive health

have attracted increasing attention. Men’s knowledge, attitudes and behaviors have been

found to have strong impacts on women’s health choices and rights. Research conducted

in Vietnam, for example, reveals that husbands are often the ultimate decision makers with

regards to the sexual and reproductive health of their wives or partners (Johansson, 1998).

Men’s knowledge and attitudes towards enabling laws and policies for women are an important

area for investigation, given their potential in influencing women’s access and use of services

and the realization of their rights. This chapter extends the discussions of male involvement

by examining men’s awareness, attitudes and behaviors around laws and policies relating to

reproductive health and gender-based violence.

7.1 Knowledge about abortion laws and services

Laws around abortion may exist, making services legal but may not be well known or utilized

due to the social stigma and cultural restrictions around abortion. The respondents in the

survey were asked a variety of questions related to abortion services, the laws and conditions

under which it was permitted. Men’s accurate knowledge and attitudes toward abortion are

considered important factors that affect their partners’ access to safe abortion services, as

well as help promote women’s reproductive health and rights. In general, abortion has been

legalized and available upon request in Vietnam since 1960 and the rights of women to request

an abortion has been highly respected (WHO, 1999). The government has issued a number of

laws that regulate abortion.

In Nepal there was a landmark constitutional change in 2002 that made abortion legal in the

country. At the time of the survey abortion had been legal in Nepal for several years and

there had been a lot of news and advocacy around the issue in the country due to recent

constitutional reforms.

Awareness of the abortion law among surveyed men in each country was relatively high. In

Vietnam, only 27 percent were not aware of any law that permitted abortion. In Nepal the

proportion was slightly higher at 36 percent (table 7.1). Only 8 percent of the men in Vietnam

and Nepal believed that any form of induced abortion is prohibited in their country. In Nepal,

only 14 percent of the men who were aware that a law existed on abortion could correctly

state the legal gestational age for an abortion, i.e. 12 weeks of pregnancy. Another 21 percent

of men said that abortion is legally permitted if a pregnancy resulted from rape and incest

and within 18 weeks of the gestational period. Among those who were aware of legal abortion

in Vietnam, a majority of them stated that an abortion is allowed if the fetus is abnormal (51

percent) and if the pregnancy causes physical or mental problems to the women.

7. MEN’S KNOWLEDGE ABOUT AND

ATTITUDE TOWARDS LAWS AND POLICIES

ON REPRODUCTIVE HEALTH AND RIGHTS

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 55

KNOWLEDGE ABOUT THE PLACE FOR SAFE ABORTION SERVICES

FIGURE 7.1

KNOWLEDGE ABOUT ABORTION AND LEGAL CONDITIONS FOR ABORTION

TABLE 7.1

Men also need to have good information on where to obtain safe abortion services in case

they need to help their wives/partners avail them. The overwhelming majority of men in both

countries believed that government hospitals are the safest places a women can access to

terminate her pregnancy (figure 7.1), followed by private clinics (38 percent in Nepal and 15

percent in Vietnam). However, a modest proportion in Vietnam (22 percent) and much smaller

one in Nepal (8 percent), did not know where these services are available.

Characteristics

Nepal Vietnam

Percentage n Percentage n

Not aware of an abortion law 35.8 358 26.5 377

Knowledge about condition for

legal abortion

If the pregnancy causes

physical or mental problems to

the pregnant women

30.1 301 39.7 566

Upto 12 weeks of pregnancy 13.9 139 9.9 141

If the fetus is abnormal 11.3 113 50.6 721

Abortion is not legal 7.6 76 8.4 120

Upto 18 weeks if incest or rape 20.6 206 6.1 87

Sex selection 1.2 12 1.4 20

Unwanted pregnancy/no desire

for additional children 10.3 67 3.8 55

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

56

ACCEPTABLE CONDITIONS FOR A WOMAN TO HAVE AN ABORTION

TABLE 7.2

7.2 Attitude towards abortion (by different socio-demographic factors)

Men’s attitude towards abortion can also directly affect their wife’s or partner’s access to

safe abortion services. If the husband has a conservative attitude towards abortion, he may

prevent his wife or partner from seeking an abortion.

In the survey, male respondents were asked whether they considered it acceptable for a

woman to terminate a pregnancy under various circumstances. Overall, the data reveal (table

7.2) a relatively high acceptance of induced abortion among men in the survey. Circumstances

that related to the well-being of the woman elicited the most support from men in both

countries (around 88 percent); additionally in Vietnam the same proportion of men were in

favor of an abortion if the fetus is abnormal (but only 59 percent in Nepal). Most of the men

also favored a termination when the pregnancy was the result of a rape (68 percent in Nepal

and 58 percent in Vietnam). The social stigma of unmarried pregnancies was apparent in the

large share of Nepali men (65 percent), who supported an abortion if the girl was unmarried

(in Vietnam the level of support was also high at 42 percent). Roughly half the Nepali men

perceive that it is acceptable for a woman to undergo an abortion when she has completed

her family size and half the Vietnamese men supported an abortion when the women has

economic difficulties and cannot provide for another child (in both these cases, the proportion

of men in support in the other country was lower). Interestingly, only a very low proportion of

men (2.5 percent in Nepal and 5 percent in Vietnam) supported abortion for a female fetus.

Conditions

Nepal Vietnam

Percentage n Percentage n

If the pregnancy could harm her

health 89.2 892 88.2 1257

She is pregnant from rape 68.2 682 58.3 831

She is unmarried and pregnant 65.2 652 42.7 609

The fetus is abnormal 59.3 593 88.0 1254

Pregnancy is unplanned or

unwanted 49.5 495 39.6 565

She has difficult economic

circumstances/hard to provide

for another child

44.0 440 48.9 697

She already has enough

children 44.0 440 32.4 461

She already has enough sons 39.8 398 25.1 358

She already has enough

daughters 38.6 386 24.7 352

The fetus is female 2.5 25 4.8 68

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 57

7.3 Knowledge about ultra sound test by wife/partner and attitude towards

sex selective abortion

In the survey, men’s knowledge of their wife/partner reproductive history and ultrasound tests

was also assessed. Since the early 1990s, ultrasonography scanning services have developed

rapidly and become a lucrative business in Vietnam. Because of its availability, accessibility

and affordability, ultrasound technology has been increasingly used by pregnant women

as part of routine pregnancy care (Gammeltoft and Nguyen, 2007). In Nepal, sex selective

abortion is not yet a major issue of concern. However in light of the legalization of abortion,

easy availability of pre-natal sex-determination technologies and abortion clinics, as well as

religious and socio-economic values given to sons, the demand for sex selective abortion could

increase in coming years (CREHPA/UNFPA, 2007b). Currently, sex selective abortion in Nepal

prevails in the urban areas of the terai region and areas neighboring India’s border (CREHPA/

UNFPA, 2007a).

Men were asked about whether or not their wife/partner had ever undergone an ultra sound

test. The results show that the majority of men in Vietnam (64 percent) noted that their

wife/partner had ever gone for an ultrasound test, while this proportion was only 17 percent

in Nepal. The lower proportion in Nepal could also be the result of the lack of access to the

technology at the scale at which it is available in Vietnam.

PARTNER EVER UNDERGONE ULTRASOUND TEST

FIGURE 7.2

An analysis of ultrasound testing revealed that many of the women had undergone more than

one ultrasound test during the last pregnancy in both countries (table 7.3). The figure was

more than 30 percent in Nepal whereas in Vietnam the majority of men (62 percent) reported

that 3 or more ultrasounds had been performed during their wife’s last pregnancy.

The majority of the respondents’ wives/partners in Vietnam had taken an obstetrical ultrasound

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam

58

STAGE OF PREGNANCY WHEN ULTRASOUND WAS DONE AND REASON FOR ULTRASOUND

TABLE 7.3

test during their latest pregnancy after the fourth month (16 weeks) of the pregnancy, while

half the men’s partners in Nepal had taken the test during the 13-28 week period. Further

examination of the reasons for ultrasound testing by gestational age revealed that, in Vietnam,

the most frequently mentioned reason was concerns about the fetus’s growth and normality

(51 percent), followed by concern for a wife’s/partner’s health (39 percent). In Nepal the most

common reasons cited for an ultrasound test were doctor’s advice (35 percent) and wife’s/

partner’s health (22 percent).

One in ten of the men in Nepal whose partner had an ultrasound test reported it was because

they wanted a son; in Vietnam this was acknowledged by 4.3 percent of the men.

Conditions Nepal Vietnam

Percentage n Percentage n

Stage of pregnancy when wife/partner had

ultrasound

2-9 weeks 12.8 16 5.1 30

10-12 weeks 24.0 30 7.7 45

13-28 weeks 52.0 65 41.6 243

More than 28 weeks 11.2 14 45.5 266

Reason for having an ultrasound

Doctor’s advice 35.2 44 1.5 10

Wife/partner health 22.4 28 38.9 255

To check fetus condition 16.8 21 - -

Wanted a son 10.4 13 4.3 28

Risk to fetus 5.6 7 51.2 336

To find whether son or daughter 4.0 5 - -

Desire for daughter 1.6 2 - -

Abortion 1.6 2 - -

To confirm pregnancy 1.6 2 - -

Family pressure 0.8 1 0.5 3

Other - - 3.7 24

Number of ultra sounds done during last

pregnancy

1 68.8 86 15.1 95

2 22.4 28 23.2 146

3 or more 8.8 11 61.7 389

Did the provider disclose the sex of the baby

himself/herself during the ultrasound

Yes 32.0 40 75.1 489

No 60.8 76 18.8 123

I was not present 5.6 7 3.6 24

Don’t know 1.6 2 2.3 15

Study on Gender, Masculinity and Son Preference in Nepal and Vietnam 59

Both Nepal and Vietnam have banned disclosing the sex of the unborn baby to parents during

ultrasound tests, yet the survey results indicate that these state regulations were not followed.

In our sample, the vast majority of Vietnamese men (75 percent) reported that the sex of the

fetus was disclosed by the service providers while undergoing an ultrasound test. This finding

echoes what has been found in the GSO 2010 Population Change and Family Planning Survey:

75 percent of women aged 15-49 who gave birth from April 2008 to March 2010 knew the sex

of the fetus before delivery. In Nepal, 32 percent of the men reported this.

According to findings from a recent qualitative study on sex ratios in Vietnam, the high

proportion of health providers disclosing the sex of the fetus to their clients was attributed to

the fierce competition among clinics, strong needs or pressures from clients to know the sex

of the fetus and the very loose enforcement of the regulations on the use of ultrasound tests

for sex identification. Private clinics in particular face greater competition in the absence of

government subsidies, and were more willing to attract clients by offering to let them know

the fetal sex which is one of the primary aims for undergoing ultrasound test (UNFPA 2011).

7.4 Men’s knowledge and attitude towards policies and laws promoting

gender equity

Over the past decade, both Nepal and Vietnam have made significant improvements in

terms of policy development and legislation to promote gender equality and women’s rights.

During this period, three important laws were passed in Nepal: there was an amendment in

the country code to legalize abortion under certain conditions, to provide property rights to

daughters, 2002 and The Domestic Violence and Punishment Act, 2009. Similarly, Vietnam has

enacted the 2006 Law on Gender Equality, the 2007 Law on Domestic Violence Prevention and

Control and the 2010 National Strategy on Gender Equality.

Another achievement on gender equality was the