Exploring The Causes and Process of Becoming Child Domestic Worker

Working Paper No.
35
Exploring the Causes and Process of

Becoming Child Domestic Worker
Shuburna Chodhuary, Md. Akramul Islam, Jesmin Akter
January 2013
Research and Evaluation Division (RED), BRAC, 75 Mohakhali, Dhaka 1212, Bangladesh
Telephone: (88-02) 9881265-72, 8824180-7 (PABX) Fax: (88-02) 8823542
E-mail: [email protected], Website: www.brac.net/research
Working Paper No. 35
Copyright © 2013 BRAC
January 2013
Page makeup and composition

Altamas Pasha
Cover design
Md. Abdur Razzaque
Design and layout

Md. Akram Hossain
Published by:
Research and Evaluation Division

BRAC
BRAC Centre
75 Mohakhali
Dhaka 1212, Bangladesh
Telephone: (88-02) 9881265-72, 8824180-7 (PABX)
Fax: (88-02) 8823542
Website: www.brac.net/research
BRAC/RED publishes research reports, scientific papers, monographs, working

papers, research compendium in Bangla (Nirjash), proceedings, manuals, and other
publications on subjects relating to poverty, social development and human rights,
health and nutrition, education, gender, environment, and governance.
Printed by BRAC Printers at Tongi, Gazipur, Bangladesh.

Table of contents
Acknowledgements v
Abstract vi
Executive Summary vii
1. Introduction 1
Background 1
Definition of CDW 2
Rational of the study 2
Study Objectives 3
2. Methods 4
Study area and population 4
Sampling procedure 4
Methods of data collection 5
Data management and analysis 5
Limitations 6
3. Findings 7
Socio-demographic profile 7
Economic empowerment Indicators 8
Health status and health services 15
Affiliation with NGO 16
Becoming child domestic workers (CDWs) 17
Condition under which parents would stop their children from engaging in
domestic work 24
Working environments of child domestic workers 25
Respondents’ knowledge on child rights 27
4. Discussion and Conclusion 31
5. Recommendations 33
References 34
Annexure 36
The causes and process of becoming child domestic worker iii

Acronyms
ADP Adolescents Development Programme

Agri. Agriculture
BBS Bangladesh Bureau of Statistics
BEP BRAC Education Programme
BHP BRAC Health Programme
CEP Community Empowerment Programme
CDW Child Domestic Worker
GD Group Discussion
HRLS Human Rights and Legal Aid Services
HSC Higher Secondary School Certificate
HIES Household Income Expenditure Survey
HH Households
IGA Income Generating Activities
ILO International Labour Organization
NGO Non- governmental Organization
NIPORT National Institute of Population Research and Training
Non-agri Non-agriculture
PRA Participatory Rural Appraisal
RED Research and Evaluation Division
SPSS Statistical Package for Social Science
SSC Secondary School Certificate
TUP Targeted Ultra Poor
UNICEF United Nations Children’s Emergency Fund
UN United Nations
iv RED Working Paper No. 35

Acknowledgements
The authors are grateful to Dr.Faustina Pereira, Director, Human Rights and Legal
Services and Dr. WMH Jaim, Director, Research and Evaluation Division, BRAC for
their support. They are also thankful to Dr. Mohammad Rafi, Head of Research (SD &
HR unit) Research and Evaluation Division, BRAC for his guidance in designing and
analysis of the study. Special thanks given to Ms. Ishita Islam, Senior Programme
Specialist and Ms. Dilshad Mahmud, Staff lawyer of BRAC, Human Rights and Legal
Services for providing continuous support. We also like to thanks field management
and field researchers of Research and Evaluation Division, who assisted in data
collection and processing. Sincere thanks to Mr. Hasan Shareef Ahmed and Mr.
Altamas Pasha for editing and final proof reading of the manuscript, respectively. Mr.
Md. Akram Hossain formatted the report for publication also deserves thanks. The
authors are also indebted to the survey respondents for their available time and
useful data for the study. Thanks to all people associated with this research for their
support, suggestions and assistance.
RED is supported by BRAC's core fund and funds from donor agencies,
organizations and governments worldwide. Current donors of BRAC and RED
include Aga Khan Foundation Canada, AusAID, Australian High Commission, Bill and
Melinda Gates Foundation, Canadian International Development Agency, CARE-
Bangladesh, Department for International Development (DFID) of UK, European
Commission, Euro consult Mott Mac Donald, Global Development Network Inc
(GDN), The Global Fund, GTZ (GTZ is now GIZ) (Germany), Government of
Bangladesh, The Hospital for Sick Children, Institute of Development Studies
(Sussex, UK), Inter-cooperation Bangladesh, International Labour Office (ILO), IRRI,
Liverpool School of Tropical Medicine, Manusher Jonno Foundation, Micro-Nutrient
Initiative, NOVIB, Plan Bangladesh, Embassy of the Kingdom of the Netherlands,
Swiss Development Cooperation, UN Women, UNHCR, UNICEF, Unilever-UK,
University of Leeds, World Bank, World Food Programme, World Fish, Winrock
International USA, Save the Children USA, Save the Children UK, Safer World,
Rockefeller Foundation, BRAC UK, BRAC USA, Oxford University, Karolinska
University, International Union for Conservation of Nature and Natural Resources
(IUCN), Emory University, Agricultural Innovation in Dryland Africa Project (AIDA), AED
ARTS, United Nations Development Program, United Nations Democracy Fund,
Family Health International, The Global Alliance for Improved Nutrition (GAIN), Sight
Saver (UK), Engender Health (USA), International Food Policy Research Institute
(IFPRI) and Yale/Stanford University.
The causes and process of becoming child domestic worker v

Abstract
The study aimed to explore the demographic and socioeconomic characteristics of

households with and without child domestic workers (CDW), and explore the causes
and process of becoming CDWs in Bangladesh. Both quantitative and qualitative
approaches were used to collect data. A purposive sampling technique was applied
to select the sample locations. Data were collected from six districts of Bangladesh:
Mymensingh, Kishoreganj, Jamalpur, Bhola, Kurigram, Chapai Nawabganj, and a
slum of Dhaka city. A total of 1,564 (rural 1,454 and urban 110) parents of with and
without CDWs were interviewed. Ten in-depth interviews with CDWs were
conducted. The findings reveals that small income, poor savings, marginal land, less
access to NGO services, and sickness of main income earner in the family were the
causes that compelled parents to allow their children in CDW. The findings also
strongly observed that poverty is the root cause to engage children in domestic work.
The findings also reveal that neighbours, relatives, employers, and middlemen as the
intermediary were effective and influential too for CDW in both the study areas.
However, parents and CDW herself/himself in urban areas were the most
instrumental factors to get into domestic work. Ensuring economic empowerment,
educational facilities, health services, and raise awareness on child rights among the
households with and without CDWs through BRAC programmes could slow down,
children involvement in domestic work.
vi RED Working Paper No. 35

Executive summary
Introduction
Children’s involvement in domestic work for income is a common phenomenon in a

poor country like Bangladesh. Such employment is a violation of child rights. Most of
these children come from the poor and extremely poor families. BRAC Human Rights
and Legal Aid Services (HRLS), one of the core programmes of BRAC, will launch a
project on reducing children’s involvement in domestic work. Before launching the
project this study has been conducted so that it may contribute in enhancing the
effectiveness of the project. The study aims to explore the demographic and
socioeconomic characteristics of households with or without Child Domestic Workers
(CDW), as well as to explore the cause and process of becoming CDWs.
Methods
Both approaches quantitative and qualitative were used. Data were collected from six
districts of Bangladesh: Mymensingh, Kishoreganj, Jamalpur, Bhola, Kurigram,
Chapai Nawabganj, and a slum of Dhaka city. A total of 1,564 (rural 1,454 and urban
110) parents of with and without CDWs were interviewed. Ten in-depth interviews
with CDWs were conducted.
Key findings
Socio-demographic profile
The study reveals that the household size in rural areas (4.8) was lower than that of
urban areas (5.2). In the observed households, 5-17 years age-group was higher
compared to other age groups in both rural and urban areas. The 18-40 year age-
group constituted the second largest population group with 35.3% in rural and
29.9% in urban areas. A significant number of observed populations had “no
education” in both areas (rural 48.3%, urban 67.3%). However, a good number of
respondents completed primary education in both rural (40.1%) and urban (26.7%)
area. A significant number was unmarried in both areas (rural 41.7%, urban 35.7%).
In terms of main occupation, the responses were aggregated into eight broad
occupational categories, e.g., agriculture, small business, service, transport worker,
day labour (agricultural, non-agricultural), fishing, domestic worker, and household
work. Among them, it was observed that majority of target population was involved in
non-skilled physical labour intensive occupations, e.g., day labourer (19.5% rural,
12.7% urban), transport worker (3.7% rural, 11.6% urban), domestic work (1.4%
rural, 16.5% urban) in both areas. Some of these activities were not directly income
generating activities in nature (non-IGA) such as household work (rural 20.9%, urban
13.1%) and a good number of people in both rural and urban areas were involved in
it.
The causes and process of becoming child domestic worker vii

Economic empowerment indicators
Four economic empowerment indicators, e.g., income, expenditure, savings, and

credit, were assessed. It was observed that the average income, expenditure, and
savings of surveyed households were significantly different between rural and urban
households. Firstly, the average annual income of households without CDWs (Tk.
79,255 in rural area and Tk. 100,546 in urban area) was slightly more than CDW
households (Tk. 71,872 rural, Tk. 93,181 urban) in both the areas. Secondly, the
average annual expenditure of observed households also followed similar trend, i.e.,
average annual expenditure of households without CDWs (Tk. 74,168 rural, Tk.
90,618 urban) was greater than that of households with CDWs (Tk. 67,786 rural, Tk.
87,620 urban) in both areas. The food and clothing constituted over 69% of
expenditures in both areas. Thirdly, the average savings of households without CDWs
(Tk. 4,837 rural, Tk. 3,013 urban) was much higher than that of households with
CDWs (Tk. 2,295 rural, Tk. 772 urban) in both areas. A good percentage of surveyed
households in rural (34% of households with CDWs and 21.5% of without CDWs)
and urban areas (44.7% of households with CDWs and 34.9% of without CDWs) had
no savings to combat any crisis. Fourthly, nearly 40% of surveyed households did not
receive any credit from any formal sector like non-government organizations (NGOs)
in the preceding year. However, the average loan size was Tk. 19,692 for households
with CDWs and Tk. 18,689 for households without CDWs in the preceding year. The
percentage of landless households with CDWs (30%) was higher than households
without CDWs (22.4%) in rural areas, while most of the urban households surveyed
had no land affected by river erosion. In the case of land ownership, it was observed
that most of the land holdings were <10 decimals, 54.0% of households with CDWs
and 62.5% of households falls under this category without CDWs in rural areas
respectively.
Health status and health services
More than 50% of the respondents suffered from illnesses during the past 14 days
(rural 55%, urban 52%). Fever, common cold, different types of pain were the most
commonly reported illnesses. Besides, about 10% of the respondents suffered from
diarrhoea and dysentery. Some people sought treatment at local drug stores, which
appeared to be the major healthcare service provider (69% of cases in the drug store
compared to 19% of cases in government hospitals) in the study areas respectively.
Becoming child domestic workers
Result shows that 14-17 years old girls (rural 40%, urban 53.8%) were engaged more
in domestic work than boys (rural 28.6%, urban 25%) in both rural and urban areas.
On the other hand boys (rural 71.4%, urban 75%) were engaged more in domestic
work than the girls (rural 46.2%, urban 60%) at the age of 6-13 years respectively.
Majority of the CDWs (rural 61%, urban 53.1%) from both areas had no education
and a moderate percentage (rural 32.2%, urban 38.8%) of the CDWs had primary
viii RED Working Paper No. 35

education. A small number (rural 6.8%, urban 6.1%) of CDWs had secondary
education respectively.
Parents compelled to engage their children in domestic work was higher in urban
areas (28.6%) compared to rural areas (1.7%). CDWs engaged in domestic work by
themselves was 14.3% in urban areas. No CDWs in rural areas were found to be
involved in domestic work by themselves. Apart from this, relative, neighbour,
employers, and middlemen were the most instrumental in getting CDWs involved in
domestic work. The parents in urban areas engaged their children in domestic work
for supporting livelihood in a city.
The study found that CDWs came to workplace with parents and by themselves
were higher in urban areas compared to rural areas. On the other hand, CDWs came
to work place with neighbour and middlemen were higher in rural areas compared to
urban areas.
Most of the respondents came to Dhaka to seek employment (34.9%), poverty

(31.8%), river erosion (15.5%), to escape from repay the loan (11.6%) and their family
need more income (2.3%).They also mentioned other causes such as broken family
and lack of land. Most of these families came from Bhola, a river erosion area of
Bangladesh.
Among the children <18 years, involvement in income generating activities (IGA) were
higher in urban areas (35.5%) compared to rural areas (17.5%). Besides, parents
wanted to engage children before reaching 18 years of age in IGA were higher in
urban areas (54.6%) compared to rural areas (32.5%), as parents from urban areas
were more interested to engage children in IGA than the parents of rural areas.
Most of the parents of CDWs mentioned that they engaged their children in domestic
work due to poverty. Besides, a few parents in rural areas had mentioned other
reasons such as repay the loan which had been borrowed to give dowry in
daughters’ wedding (3.4%), fathers’ sickness (1.7%) and CDWs father did not look
after the family (1.7%). Parents from urban areas mentioned other causes such as
large family size (4.1%), family needs more income (4%), and savings money for
CDWs wedding (2%).
Of the study areas, most of the CDWs came to Dhaka from Mymensingh (36.4%),
Bhola (30.3%), Kishoreganj (18.2%), and Kurigram (12.1%). Only a negligible
percentage of CDWs came to Dhaka from Jamalpur (3.1%).
domestic work
Most of the CDWs’ parents mentioned that if their family has sufficient income they
would not have engaged their children in domestic work. A few of the CDWs’ parents
from rural areas had responded other conditions, such as if family could arrange
money for CDWs marriage (5.1%), repay the loan (2.5%), earning member health
The causes and process of becoming child domestic worker ix

condition improved (2.5%), and family could ensure security for CDWs (1.3%), the
family would withdraw their children from work. Parents from urban areas also
mentioned that if their family could arrange money for CDWs marriage (3.8%), they
would not engage them in domestic work. The parents also mentioned other
conditions such as if they could arrange money to build new house in their village
(7.7%) and if their children became adult they would engage them in other
professions instead of domestic work (7.6%).
Working environments of child domestic workers
About 40% of CDWs from rural areas and 85% from urban areas received food,
lodging and clothing with salary from employers. It means CDWs from rural areas
received less benefits from the employers compared to CDWs from urban areas.
Distribution of CDWs salary also indicates that CDWs from urban area were paid
more than CDWs from rural areas. Similar distribution had been observed of CDWs
sending money to parents. Most of the CDWs in urban and rural areas were paid Tk.
501-1,000 as a salary.
A little more than half of the CDWs’ parents from both areas said that their children
did not mention about any problem faced at work place. Most of those who
mentioned about problems faced at work place, mentioned about too much work
load, scold/verbal abuse and physical assault. A few parents from both areas
mentioned about other problem faced at work place such as less food provided, did
not allow children to go home, and low wage.
Respondents’ knowledge on child rights
According to UN convention of child rights and based on our definition of child age
we considered less than 18 years of age an individual is called child. It was also
considered that at 18 years of age an individual became earning person. Among the
respondents who could answer the question according to our definition, we
considered that respondents had knowledge on child age and working age of an
individual. The findings show that most of the respondents both in urban and rural
areas with and without CDWs had no knowledge about child age. Although,
knowledge on child age with and without CDWs found statistically significant within
rural areas. In contrast, there was no significant difference was found with and
without CDWs within urban areas. Almost similar findings were observed about the
knowledge on earning age of an individual.
x RED Working Paper No. 35

Recommendations
In any civilized society the system of child labour is not acceptable. Furthermore, it is
strongly prohibited under International Labour Organization (ILO) convention on the
elimination of child labour. Therefore, GO, NGO like BRAC can take programmes or
initiative to slow down children’s involvement in domestic work. The study in
particularly created a data bank consisting demographic profile of households with
and without CDWs with location. It is hence anticipated that by using this data bank,
BRAC can extend its services to these households by providing support of economic
empowerment, health, education and create awareness to slow down children
involvement in domestic work.
The causes and process of becoming child domestic worker xi

1. Introduction
Background
Children’s involvement in domestic work is an invisible and cheap form of

employment as well as widely practiced across the globe. For example,
approximately 175,000 children aged <18 years in Central America, more than
688,000 in Indonesia, 76600 in Philippines, 53,942 in South Africa and 38,000 in
Guatemala are engaged in domestic work. ILO defines children’s involvement in
domestic work as one of the worst forms of child labour as because children involved
in domestic work were deprived from familial affection, education, healthcare, and
recreation (www.ilo.org, Black 1996). Many studies indicate that children involvement
in domestic work has a negative impact of their psychological and physical
development (Kielland and Tovo 2006; Black 1996; Bourdillon et al. 2010;
Hawamdeh and Spencer 2003).
Bangladesh is a densely populated country with 142 million people (BBS 2011).
According to child labour survey in 2003, 7.5% of children in Bangladesh aged
between 5-17 years were involved in labour market (BBS 2003). Out of them,
4,21,426 children under 18 were involved in domestic work. Again among them,
147,943 child domestic workers (CDW) were employed in Dhaka. Most of the CDWs
in Bangladesh were female (78%) (BBS and UNICEF 2006). About 60% of the CDWs
faced variety of abuses, e.g., verbal, physical and sexual. More than half of the CDWs
were compensated with accommodation, clothing and food without giving salary for
their work by the employers (BBS and UNICEF 2003). Despite this exploitation, the
Labour Law Policy 2006, do not have any clause for the protection of domestic
workers. This is because the domestic work is considered as an informal job in
Bangladesh (Save the Children 2009). Usually parents in rural areas with extreme
poverty forced their children in IGA including domestic work. Because, poor parents
are more likely to spend this source of income in meeting family expenses (Bourdillon
et al. 2010). Poverty is the main cause of children’s involvement in domestic work. A
study shows that poor parents had the greatest influence in sending their children in
domestic work (Save the Children 2009). Approximately 25.1% of people in
Bangladesh lived below the poverty line, among them 28.6% lived in rural areas and
14.6% lived in urban areas (BBS 2012). So, the present situation suggests that the
households below the poverty line are likely to involve their children in IGA including
domestic work.
BRAC HRLS programme started its journey in 1986 by providing legal aid services to
poor and disadvantaged women. Now this programme initiates a project to prevent
children in engaging domestic work. So, this explorative study had been conducted
to know the causes and process of CDW as well as to create a data bank consisting
demographic profile including location of households with and without CDWs.
The causes and process of becoming child domestic worker 1

Definition of CDW
Child Domestic Worker: CDW is one aged between 5-17 years moved out of the
own village to another area since one month for domestic work such as cooking,
washing dishes and clothes, cleaning the house, looking after employers’ children,
and any other household activities as suggested by the employers. The CDWs mainly
come from rural households with extreme poverty and resided in employers’
households. For their services they are compensated and that comes in different
forms like lodging, boarding, cloths, salary, etc.
Rational of the study
Children involvement in labour market firstly revealed in 1996 in BBS national survey
on child labour. According to this survey, 6.3 million children aged 5-14 years were
involved in labour market. To update the knowledge of child labour a follow up survey
has been conducted nationwide by BBS in 2003 (BBS 2003). This survey indicates
that 1.3 million children are involved in worst form of labour, defined as children
working more than 43 hours or more in a week. A study in Save the Children showed
that more than half of the children work 9 to 15 hours in a day (Save the Children
2009). According to this definition child involvement in domestic work is worst form of
labour.
Child labourers in other informal sector are more visible than the CDW. Since they
lived in employers’ house and hidden from public view. This forms of labour violates
the convention of child rights (CRC) in 1990. According to the articles of CRC a child
has right to enjoy non-discrimination, cared for by parents, preserve identity,
nationality, name and family relation, freedom of express his/her own views,
thoughts, religion, protection from abuse and neglect, access to healthcare,
recreation and leisure time, and protection from sexual exploitation. Bangladesh was
one of the countries who ratify the above mentioned convention (BBS 2005). Despite
this signatory, the above mentioned statistics in Bangladesh showed that children
involvement in domestic work is a great matter of concern.
However, culturally children engagement in domestic work is accepted and practiced

across Bangladesh. Traditionally parents trained their daughter children in household
task. Today this traditional help is being commercialized due to rapid urbanization
(Black 1996). CDWs survey in 2006 indicates that CDWs concentration was higher in
large municipalities compared to rural areas (BBS and UNICEF 2006). Rahman in his
qualitative study entitled “The Child Labour Situation in Bangladesh” identified “push
factors” “pull factors” and “interactive factors” that compel children to involve in
domestic work. Child domestic worker survey in 2006 also mentioned that more than
80% of CDWs comes from poor family (BBS and UNICEF 2006). Number of study
across the globe also mentioned that parental socioeconomic situation compel
parents to involve children in labour force including domestic work (Burra N 1997,
Dixit 1997; Elgbeleye and Olasupo 2011). Despite this, limited research has been
conducted to explore the socioeconomic characteristics of poor and extreme poor
households.
2 RED Working Paper No. 35

It is assumed that the households below the poverty line irrespective of that
household with CDWs or not, have the possibility to involve their children in labour
force including domestic work. Thus, a study on CDW should first identify this group.
Exploring the process of becoming CDW and its multifaceted causes that compel
parents to put children in domestic work can assist in developing interventions to
reduce the risk to put children in domestic work. This exploratory study aims to
investigate the process and root causes that force children to become CDW. The
study also created a data bank containing the socioeconomic and demographic
information of households with and without CDWs with location. The findings of the
study will assist the BRAC HRLS programme to take a new project on CDW. The
new project will be implemented with an integrated approach of BRAC development
initiatives by targeting ultra poor to reduce the rate of child domestic work among the
poor and extreme poor families.
Study objectives
• Create a data bank containing demographic information on parents, families and

communities with and without CDWs and the location.
• To explore the cause that force children to become CDWs.

2. Methods
Study area and population
The study was conducted in six districts from four divisions along with one urban
slum from Dhaka in Bangladesh. One district from a division was selected, except
Dhaka division where three districts were selected for the study. It is assume that
these three districts have high concentration of the CDWs. The study site and
population are mentioned in Table 1.
Table 1. Study site
Division District Upazila/Area Number of Number of

Village/Slum HH
Dhaka Mymensingh Shambugonj 3 227
Kishoreganj Katiuadi 3 309
Jamalpur Jamalpur sadar 3 205
Dhaka City Mohammadpur (Basila) 1 110
Rangpur Kurigram Kurigram sadar 3 241
Rajshahi Chapai Nawabganj Chapai Nawabganj sadar 3 216
Barisal Bhola Bhola sadar 3 256
4 divisions 6 districts and one city 6 upazilas and one slum 18 villages and 1,564
one slum
Sampling procedure
A purposive sampling technique was applied to select the sample locations. Firstly,
six districts were selected from the four divisions (Dhaka, Rangpur, Barisal and
Rajshahi) of Bangladesh positioned at lower level in terms of selected development
indicators, e.g., poverty ratio, literacy rate, school attendance rate of children aged
>5 years. Secondly, six upazilas were selected from each of the six districts that were
observed to have maximum number of BRAC programmes such as adolescents
development programme, education, targeted ultra poor (TUP) and health. Thirdly,
after consulting BRAC staff, volunteers and local leaders, the field workers selected
18 villages, three villages from each six upazilas considered to be least developed
and poverty prone areas. Simultaneously, the study also selected one slum which
was Bashila bustee at Mohammadpur in Dhaka city under TUP programme of BRAC
(Annexure 1).

The respondent and selected households
In each village the field workers conducted a Group Discussion (GD) with the
participation of villagers to prepare the village social map (Annexure 2) and conducted
wealth ranking exercise for stratifying their community into four groups or classes
e.g., rich, middle, poor and extreme poor based on their livelihood resources
(Annexure 3). One GD session in each village was conducted using a guideline
coherent with study objectives. The exercise extracted information on issues on
household, occupation, income or financial status, amount of land and number of
income earner, etc. After the session, the field workers produced a household list of
respective village including all four groups. Finally, the study selected all of the poor
and extreme poor households who had 5-17 years of children for interview. In urban
areas a door to door visit was conducted to collect information on households who
had 5-17 years children and also considered with and without CDWs.
Methods of data collection
Both quantitative and qualitative methods were used for data collection. A semi-
structured interview questionnaire was used to gather information on selected
households. We also conducted in-depth interviews with CDWs in urban area to
explore the deepen of CDWs livelihoods from individuals. Twenty trained field
enumerators assisted in data collection.
Survey
A total of 1,564 households (rural 1,454 and urban 110) comprising households with
and without CDWs were surveyed both in urban and rural areas. Survey
questionnaire comprised of closed and open-ended questions. The survey was
carried out to collect socio-demographic information of CDWs or without CDWs
families and their relationship with BRAC. We also collected information about the
process, casual explanation, preventive measures, and working environment of
CDWs.
In-depth interview
Ten in-depth interviews with CDWs were carried out to collect data on the process of
becoming CDWs, his/her migration process to Dhaka, casual explanation of
becoming CDWs, working environment, preventive measures of becoming CDWs
and CDWs knowledge on child rights.
Data management and analysis
The quantitative data were analyzed by using SPSS. Simple statistical techniques
include simple frequency distribution expressed in bivariate table with T-test.
Differences between urban and rural, with and without CDWs households were also
explored (Annexure 4). The qualitative data were analyzed manually by following
coding and recoding process. In addition, qualitative techniques were employed to

describe any particular situation of significance, understand and complement the
associations observed in quantitative finding(s).
Extensive guidance was provided to field workers by supervisor, researcher and data
management team of RED, BRAC. A number of randomly sampled re-interviews
were carried out to cross-check the reliability and validity of data.
Limitations
Domestic work is informal job in Bangladesh. However, the study population did not
recognize the domestic work as a formal job. On the other hand, employers also
treated their domestic maid as a family member and also addressing them as a
distant relative. Therefore, identifying domestic worker in study area was a challenge
for this study. We also faced challenge to conduct in-depth interviews with CDWs in
their employers’ house.

3. Findings
Socio-demographic profile
In considering the first objective of the study the socio-demographic characteristics

focused on households’ age, education, marital status and main occupation of the
households’ members. Table 2 reports summary statistics on these parameters
obtained the study for rural and urban respondents. It was observed that there were
statistically significant difference between rural and urban respondents in terms of
average size of household, income, expenditure and savings, primary and junior level
education completion rate, and marital status.
Table 2 also compares the age distribution of rural and urban household members.
On an average urban households had slightly higher number of members aged <5
years and 5 -17 years age-group compared to their rural counterpart, but the
difference was not statistically significant.
On the other hand, the average rural household had little higher number of members
in 18 - 40 years and above 60 years age-group compared to their urban counterpart,
but the difference was not statistically significant. Equal number of member
distribution was observed in 41 - 60 years of age in both the areas (rural and urban).
More than half of the population was married in both areas (rural 54.7%, urban
59.6%). Percentage of unmarried population of rural areas (41.7%) was a little higher
than that of urban areas (35.7%), and a few of the respondents were widowed and
divorced/separated in both areas.
The majority of target population was involved in non-skilled intensive physical labour
occupations, e.g., day labourer, transport worker, domestic work in both areas.
Some of these activities were not directly income generating in nature (non-IGA) such
as household work; and a good number of people in both areas (rural 20.9%, urban
13.1%) were involved in it. In addition, a notable proportion of population (rural 12%
and urban 12.9%) was involved in other occupations, e.g., begging, tailoring/katha
sewing, carpenter (work with bamboo, cane), handicraft, Kabiraz, and Polly doctor.
The education rate of rural area was better compared to urban area. However, a
significant proportion of population had no education in both the areas (rural 48.3%,
urban 67.3%). Besides, a good number of surveyed population had completed
primary education in both areas, rural (40.1%) and urban (26.7%) and the difference
was statistically significant. Less than 10% of population completed their junior level
education in both rural (7.8%) and urban (4.5%) areas, and the difference was
statistically significant. A little number of surveyed population had completed
secondary education (rural 3.4%, urban 1.8%) and the difference was not statistically
significant.

Table 2. Socio-demographic characteristics of surveyed household
Indicators Rural Urban Difference p value

(n=1,454) (n=110)
Household
Household size 4.8 5.2 0.3 .007
Annual income (Tk.) 79,001 97,399 18397.9 .000
Annual expenditure (Tk.) 74,005 89,337 15332.8 .000
Savings (Tk.) 4,750 2,056 2694.1 .033
Age (%)
<5 10.2 12.2 -.02 .127
5-17 40.7 44.8 -.04 .057
18-40 35.3 29.9 .05 .099
41-60 11.3 11.7 -.00 .779
60> 2.4 1.4 .01 .116
Education (year, %)
No education 48.3 67.3 -.19 .000
1-5 40.1 26.7 .13 .000
6-8 7.8 4.2 .03 .003
9-10 3.4 1.8 .01 .053
11-12 0.4 0.0 .00 .175
13 and above 0.1 0.0 .00 .545
Marital status (%)
Unmarried 41.7 35.7 .06 .004
Married 54.7 59.6 -.04 .023
Widow/widower 2.5 2.6 -.00 .903
Divorced/abandoned 1.0 2.1 -.01 .013
Main occupation (%)
Agriculture 0.4 -
Small business 3.4 4.8
Service 2.5 4.6
Transport worker 3.7 11.6
Day labour (Agriculture) 4.4 -
Day labour (Non-agriculture) 15.1 12.7
Fishing 4.0 -
Domestic worker 1.4 16.5
Household work 20.9 13.1
Students 30.5 21.9
Unemployed 1.7 1.8
Others 12.0 12.9
Economic empowerment indicators
Four economic empowerment indicators, e.g., income, expenditure, savings and

credit were analyzed in the survey areas which are discussed bellow.

Household income and major sources
Table 3 shows that income of households without CDWs was substantially better
than that of households with CDWs in both areas. The households without CDWs
annual average income was Tk. 79,255 in rural areas and Tk. 100,546 in urban
areas. In contrast, households with CDWs annual average income was Tk. 71,872 in
rural areas and Tk. 93,181 in urban areas. Noteably, the differences were statistically
insignificant.
Table 3. Distribution of household by income
Annual income Rural Urban p value

With CDW Without CDW With CDW Without CDW 1 vs 2 3 vs 4
(n=50) (n=1,404) (n=47) (n=63)
1 2 3 4
Average income (Tk.) 71,872 79,255 93,181 100,546 .270 .294
Income distribution (%)
<25,000 8.0 3.9 0.0 1.6 .151 .390
25,001-50,000 20.0 19.3 8.5 4.8 .902 .430
50,001-75,000 28.0 33.0 19.1 23.8 .462 .562
75,001-100,000 30.0 22.6 36.2 19.0 .220 .044
100,001-125,000 6.0 10.7 21.3 30.2 .289 .300
125,001> 8.0 10.5 14.9 20.6 .564 .445
Table 3 also shows the distribution of yearly income of households with and without
CDWs in both areas. The highest proportion of respondents (30.0%) with income
distribution between Tk. 75,001-100,000 was observed among the households with
CDWs in rural areas. On the other hand, 36.2% of the respondents with income
between Tk. 100,000-125,000 was observed among the households with CDWs in
urban areas. The number of respondents with income between Tk. 100,000-125,000
and Tk. 125,001 were observed more in urban than rural areas. It is also evident from
Table 3 that the distribution of annual income of households with and without CDWs
in urban areas showed a better situation compared to rural areas.
Figure 1. Major sources of household income in a year
Major sources (%)

100
90
80
70
60
50
Rural
40
37
30.7 Urban
30 23.6 22.4
20 12.6 11.8
14.8 13
8.5 7.9 6.7
10 4.1
1.4 0 0 0
0
Agriculture Small Service Transport Day labour Day labour Domestic Fishing
Business worker (Agri) (Non-Agri) work

Diversification of income sources played a key role to raise households’ income and
capacity to cope with food security and any crisis. Figure 1 shows diversification of
income sources in rural and urban areas. It was observed that non-agricultural day
labour (37%), agricultural day lobour (14.8%), fishing (13.0%) and transport workers
(11.8%) were more prevalent in rural areas. In contrast, domestic work (30.7%),
transport workers (23.6%) and non-agricultural day lobour (22.4%) were major
sources of income in urban areas. Besides, a certain proportion of households’
income was coming from services mainly related to garments sector (12.6 %) in
urban areas and service (6.7%), i.e., teaching or coaching in rural areas. Non-farm
self-employment like small business appears to be a reliable source of income, which
accounts for sizable income in both rural (8.5%) and urban (7.9%). It was observed
also that farm-related income sources like agricultural work (1.4%) were less
dominated in rural areas.
Expenditure and major sources of household expenditure
Household expenditures were estimated based on the average expenditure of

household head and other household members in last one year. Table 4 shows that
the annual average expenditure of households with and without CDWs was more or
less same in both areas. The average annual expenditure of households without
CDWs was Tk. 90,618 in urban areas, while annual average expenditure of the same
group was observed to be Tk. 74,168 in rural areas. Similarly, in the rural areas
households with CDWs annual average expenditure was Tk. 67,786, while average
yearly expenditure of households with CDWs was Tk. 87,620 in urban areas. The
differences were statistically insignificant (Table 4).
Table 4. Distribution of household by expenditure
Annual expenditure Rural Urban p value

(n=50 ) (n=14,04) (n=47) (n=63)
1 2 3 4
Average expenditure (Tk.) 67,786 74,168 87,620 90,618 .179 .512
Expenditure distribution (%)
< 25,000 6.0 1.9 - - .046 -
25,001-50,000 30.0 19.6 4.3 1.6 .070 .400
50,001-75,000 26.0 38.5 34.0 23.8 .075 .242
75,001-100,000 24.4 24.9 29.8 33.3 .881 .696
100,001-125,000 6.0 9.0 25.5 34.9 .468 .296
125,001> 8.0 6.1 6.4 6.3 .589 .994
Table 4 shows the distribution of yearly expenditure of households with and without
CDWs in both, rural and urban areas. Annual household expenditures in both the
areas were little higher compared to rural households’ expenditure for both
households with and without CDWs. However, an annual expenditure of <Tk. 25,000
was observed in rural areas, while this was not the case in urban areas. Among the
CDWs, 30% of the households in rural areas had annual expenditure between Tk.
25,001-50,000 a year, which was much higher than rural households without CDWs

following the same expenditure range. The difference was statistically insignificant.
On the other hand, among the households without CDWs, 38.5% of the households
in rural areas had annual expenditure between Tk. 50,001-75,000 a year, which was
a little higher than rural households with CDWs following the same expenditure range
and the difference was not statistically significant.
In contrast, among the households without CDWs, 34% of the households in urban
areas had annual expenditure between Tk. 50,001-75,000 a year, which was higher
than urban households with CDWs following the same expenditure range and
difference was statistically insignificant. Simultaneously, among the households
without CDWs, 30% of households in urban areas had annual expenditure between
Tk. 100,001-125,000 a year, which was higher than urban households with CDWs
following the same expenditure range. The difference was statistically insignificant.
Among the households with and without CDWs around 6% of the households had
annual expenditure above Tk. 125,001 in both, rural and urban areas and the
differences were statistically insignificant.
Figure 2. Major items of household expenditure in a year
100 Major items (%)

90
Rural Urban
80
69.8 69.3
70
60
50
40
30
20
10.1 8.5
4.5 4.8 6.7
10 3.1 2.3 2.8 2.6 2.2 3.6 1.5 2.6 2.8 1.1 1.8
0
Communicatio
Treatment
repairing/tent
Entertainment
Education
Investment
repayment
Fuel
Food and
clothing
n/mobility
Loan
House
Table 5. Major sources of household expenditure in a year (Tk.)
Sources Rural Urban p value

(n=50) (n=1,404) (n=47) (n=63)
1 2 3 4
Food and clothing 48,023 48,104 62,252 57,048 .972 .089
House repairing /rent 3,579 3,065 7,925 9,402 .754 .180
Loan repayment 5,280 6,428 1,696 2,754 .362 .329
Fuel 1,800 2,046 5,341 6,166 .516 .367
Treatment 2,746 3,379 2,560 2,790 .498 .632
Education 950 2,652 947 2,885 .018 .062
Communication/mobility 1,292 1,822 2,427 2,349 .182 .892
Investment 1,450 3,470 1,003 1,595 .310 .456
Entertainment 451 1,121 863 2,178 .054 .001

Different expenditure patterns between the households with and without CDWs
emerged when we looked at different components of household expenditure (Table 5
and Fig. 2). Expenditures on food and clothing constituted over 69% of total
expenses in both rural and urban areas. The second highest household expenditure
on house repairing/renting was rather very low compared to expenditure on food and
clothing in both areas. On average 6.7% of urban and 2.8% of rural households
spent on fuel which were mainly gas and electricity in urban areas and firewood,
cow-dung, leaves, etc. in rural areas. Among the non-food items, households spent a
small proportion on loan which was 8.5% in rural areas and 2.6% in urban areas. The
households from both areas spent a little proportion of their money (rural 2.3%, urban
2.8%) for communication/mobility, e.g. bus, rickshaw, van, or boat fares. Costs for
healthcare or treatment for both types of households were also part of their annual
expenses, 4.5% in rural and 3.1% urban areas. Households investment, like in small
businesses or land leasing, were also part of annual expenditure i.e. in rural areas
3.6% and 1.5% in urban areas. Money spent for education and entertainment were
very small in both areas. A significant amount of expenditure on entertainment in
households without CDWs was higher compared to that in households with CDWs in
urban areas (Table 5).
Savings
Household savings were observed in terms of average savings and its distribution by
households with and without CDWs in both, rural and urban areas. Savings is vital for
forming financial capital and coping with unexpected crisis. Analysis of savings
behaviour reveals that savings pattern/trend of households without CDWs was better
compared to households with CDWs in both the areas (Table 6).
Table 6. Distribution of household by savings
Savings Rural Urban p value

With CDW Without With CDW Without CDW 1 vs 2 3 vs 4
(n=50) CDW (n=47) (n=63)
(n=1404)
1 2 3 4
Average savings (Tk.) 2,295 4,837 772 3,013 .179 .035
Savings distribution (%)
No savings 34.0 21.5 44.7 34.9 .036 .304
<1,000 28.0 19.9 40.4 25.4 .159 .096
1,001 – 5,000 22.0 36.4 10.6 23.8 .037 .078
5,001 – 10,000 10.0 12.9 4.3 7.9 .548 .439
10,001 -15,000 2.0 3.6 0.0 4.8 .556 .132
15,000> 4.0 5.8 0.0 3.2 .596 .221
Table 6 also shows that average savings of households without CDWs were much
higher than households with CDWs in both areas. However, a good proportion of
households had no savings in both the areas, and the differences were not
statistically significant.

Among the urban households without CDWs 40.4% had average savings <Tk. 1,000
which was much higher than urban households with CDWs following the same
savings range. The difference was statistically insignificant. On the other hand,
among the rural households without CDWs 36.4% had average savings between Tk.
1,001-5,000 which was little higher than rural households with CDWs following the
same savings range. The difference was not statistically significant. However, among
the rural households without CDWs 5.8% had average savings above Tk. 15,000
which was little higher than rural households with CDWs following the same savings
range, and the difference was statistically insignificant. None of the households with
CDWs in urban areas had savings between Tk. 10,001-15,000 and above Tk. 15,000
which was more than 3% of households without CDWs following the same savings
range (Table 6).
Credit or loan
Access to credit was crucial for coping with unexpected crisis and smooth
consumption in bad times. It was also important for capital formation to expand or
initiate IGA.
Table 7. Amount of loan received by household in last year
Loan Rural p value

With CDW Without CDW
(n=50) (n=1,404)
Average loan size (Tk.) 19,692 18,687 .818
Loan distribution (%)
No loan received 48.0 36.2 .088
1,000-5,000 6.0 7.1 .775
5,001-10,000 20.0 18.0 .721
10,001-15,000 4.0 13.2 .056
15,001> 22.0 25.5 .577
Table 7 represents the amount of loan received by both types of households in the
last year. The average loan size for households with CDWs was slightly higher than
households without CDWs, and the difference was not statistically significant. Results
show that greater number of households with CDWs did not receive any loan
compared to that of households without CDWs. The number of households took loan
varied widely across class interval for both with and without CDWs. However, the
differences were not statistically significant.

Figure 3. Distribution of household by sources of loan taken in last year (%)
100 Sources of loan (%)

90
80
70
60
50
40
26.3
30 21.2 20.5
14.2 13.6
20
10 2.9 1.4
0
Brac Grameen Bank ASA Other NGOs Relatives/friends Mohajan Krishi Bank
Figure 3 shows that a little higher proportion of household members took loan from
Grameen Bank (21.2%), followed by ASA (20.5%) and BRAC (14.2%). However,
more than 25% of the surveyed household members took loan from other local
NGOs in the last year. The surveyed households also took loan from informal sectors,
e.g., relatives/friends (13.6%) and Mohajan (2.9%) to meet their crisis. It also appears
that a few households went to Krishi Bank for loan.
Ownership of land
Historically in Bangladesh ‘land poor’ are the poor in general and there have always
been a strong negative correlation between land ownership and incidence of poverty
(BBS 2007).
Table 8. Distribution of household by land ownership
Land owned Rural p value

(in decimal) With CDW Without CDW
(n=50) (n=1,404)
Own land 6.38 7.53 .596
Own agricultural land 2.34 2.56 .893
land leased 5.0 7.84 .481
Land distribution (decimal, %)
No land 30.0 22.4 .210
<10 54.0 62.5 .221
11-50 12.0 12.3 .946
51 > 4.0 2.7 .583
Table 8 shows that the proportion of land ownership of households without CDWs is
slightly high compared to households with CDWs. However, among the rural
households with CDWs 30% did not have any land which is little higher than
households without CDWs within the same land range. It was found that the highest
percentage of land holders both for with CDWs (54.0%) and without CDWs (62.5%)
belong to less than 10 decimals land holding groups. However, the differences were
statistically insignificant.

Health status and health services
To analyze the prevalence of disease, the respondents were asked to recall their
illness during the last 14 days from the date of interview. The prevalence of disease
was 55% in rural population and 52% in urban population (Fig. 4).
Figure 4. Illness status of household members in last 14 days (%)
Became ill (%)

100
80
55 52
60
40
20
0
Rural Urban
Table 9 reports that in rural areas more households with CDWs suffered from
different illness compared to households without CDWs. It was observed that fever,
common cold and different types of pain were the three most common illnesses the
respondents suffered in the last 14 days in both the areas. A substantial proportion of
the households suffered from waterborne diseases like diarrhoea and dysentery in
both the areas.
Table 9 Type of illness and healthcare seeking behaviour of household

members (%)
Types of illness Rural Urban

With CDW Without CDW With CDW Without CDW
(n= 35) (n=718 ) (n=28 ) (n=53 )
Fever/common cold 71.4 67.5 60.7 62.3
Diarrhea/dysentery 17.1 9.6 14.3 15.1
Abdominal pain 17.1 7.9 - -
Body pain 20.0 19.2 25.0 15.1
Asthma 8.6 4.9 3.6 3.8
Others 20.0 20.6 14.3 11.3
Source of treatment
Govt. Hospital 31.5 29.6 10.7 5.7
NGO clinic 2.9 2.5 - -
Brac Sasthya Kormi 14.3 29.4 - -
Drug shop 65.7 48.9 78.6 84.9
Others (Homeopathy, 2.9 7.2 3.6 3.8
Kabiraji, Jhar-fook)
No treatment 2.9 6.7 7.1 5.7
Multiple responses

Majority of the number of households with and without CDWs went to the two most
common sources for treatment i.e., local drug stores and government hospitals in
both rural and urban areas. On the other hand, seeking treatment from homeopathy
doctors and Kabiraj was a little prominent among the households without CDWs
compared to households with CDWs in both the areas. In rural areas, more from
households without CDWs (29%) sought treatment from BRAC Shasthya Kormi
compared to households with CDWs. However, there was no NGO clinic and
Shasthya Kormi in urban areas. In addition, a little proportion of the respondents,
especially in the urban areas either sought no treatment or self-treatment (Table 9).
Affiliation with NGOs
Table 10 shows that more households without CDWs were affiliated with different
NGOs i.e., BRAC, Grameen Bank, ASA compared to households with CDWs.
However, ninety one percent of households with CDWs were not affiliated with any
NGO while it was 86% for households without CDWs. Similarly, most of the
households with and without CDWs were not affiliated with any NGO in urban areas.
However, in urban areas there were some local NGOs operating but in a very small
scale.
Table 10. Percentage of household members affiliated with NGOs
Name of NGOs Rural Urban

(n=50) (n=1404) (n=47) (n=63)
BRAC 1.4 2.8 - -
Grameen Bank 2.9 4.1 - -
ASA 2.2 4.0 0.0 0.8
Other NGOs 5.0 5.3 1.7 2.3
No affiliation 91.0 85.9 98.3 97.0
Multiple responses
Table 11. Percentage of household members affiliated with programme of

BRAC
Programme Rural Urban

(n=50) (n=1,404) (n= 47) (n=63)
BRAC Education Programme (BEP) 4.3 7.4 2.6 2.6
BRAC Micro-finance Programme(BMP) 1.7 3.7 .0 .7
BRAC Health Programme (BHP) 1.7 0.8 .0 .3
BRAC Agriculture and food security 0.3 0.1 - -
programme
Challenging the Frontiers of Poverty 0.3 0.1 0.4 .0
Reduction- Targeting the Ultra Poor
(CFPR-TUP)
BRAC Community Empowerment .0 0.1 .0 .0
Programme (CEP)
Non-beneficiary 92.7 88.1 97 96.4
Multiple responses

Table 11 presents the household members’ affiliation with different BRAC
programmes. The majority of the households were involved with BRAC Education
Programme (BEP) in both the areas. In rural areas, more members of households
without CDWs tend to send their children to BRAC schools compared to households
with CDWs. In urban areas, the tendency to involve with BEP was same (2.6%) in
both types of households. Households without CDWs had took more loans from
BRAC compared to households with CDWs. In rural areas, households with CDWs
sought more treatment from BRAC Health Programme (BHP) compared to
households without CDWs (Table 10). The same scenario appears in the case of
Targeting the Ultra Poor (TUP) and Community Empowerment Programme (CEP)
programme (BRAC 2009).
Becoming child domestic workers (CDWs)
This section discusses about the background of CDWs and the process how a child
becomes a CDW. A total of 59 CDWs in rural areas and 49 CDWs in urban areas
were identified based on the definition of CDW. In terms of age CDWs were
categorized into three groups. The survey result shows that the girls aged between
14-17 years (rural 53.8%, urban 40%) were engaged more in domestic work than
boys (rural 28.6%, urban 25%) both in urban and rural areas. On the other hand,
boys aged (rural 14.3%, urban 25%) 6-9 years were engaged more in domestic work
than the girls (rural 7.7%, urban 17.8%). Similar distribution had been found in the
age group 10-13 years (Table 12).
Table 12. Percentage distribution of CDWs by age and sex (%)
Age group Rural Urban

(year) (n=50) (n=47)
Boys Girls Total Boys Girls Total
6-9 14.3 7.7 8.5 25.0 17.8 18.4
10-13 57.1 38.5 40.7 50.0 42.2 42.9
14-17 28.6 53.8 50.8 25.0 40.0 38.8
About 60% of CDWs in rural areas and 53.1% of CDWs in urban areas had no
education. A little more than one-third of CDWs both in urban and rural areas had
primary education against 6.1% of CDWs in urban areas and 6.8% in rural areas had
secondary education (Fig.5).
Table 13 reveals that parents engaged their children to get involved in domestic work
was significantly higher in urban areas (28.6%) compared to rural areas (1.7%).
CDWs’ involvements in domestic work by themselves were also significantly different
between urban and rural areas. Apart from this, relative, neighbour, employers and
middleman were the most influential persons who offered CDWs to involve in
domestic work. A small number of CDWs from both areas were proposed domestic
work by their siblings.

Figure 5. Percentage distribution of CDWs by schooling
Educational status of CDWs (%)

100
90
80
70 61 Rural Urban
60 53.1
50
38.8
40 32.2
30
20
6.8 6.1
10 0 2
0
No education Primary Secondary Religious education
Table 13. Person proposed/offered domestic work to CDWs (%)
Person who offered Rural Urban p value

(n= 50) (n= 47)
Relative 37.3 20.4 .057
Neighbour 37.3 22.4 .097
Employer 10.2 2.0 .089
Sibling 3.4 4.1 .851
Parent 1.7 28.6 .000
Middleman 10.2 8.2 .723
Self 0.0 14.3 .002
CDWs came to workplace with their parents significantly higher in urban areas
(40.3%) compared to rural areas (15.3%). In contrast, CDWs from rural areas coming
to work place with neighbours and middlemen were significantly higher compared to
those coming from urban areas. CDWs came to work place by themselves between
urban and rural areas were statistically significant. Other than the difference between
urban and rural areas of becoming CDW with the help of employers, relatives and
siblings was statistically insignificant (Table 14).
Table 14. Person with whom CDWs came to work place (%)
Person Rural Urban p value

(n= 50) (n= 47)
Neighbour 39.0 16.3 .009
Parent 15.3 40.8 .003
Middleman 20.3 0.0 .001
Employer 13.6 6.1 .207
Relative 10.2 16.3 .348
Sibling 1.7 6.1 .229
Self .0 14.3 .002

The study reveals that lack of employment opportunity have caused a stiff move to
Dhaka (34.9%) coupled with poverty (31.8%), river erosion (15.5%), debt (11.6%) and
financial need for the family for 2.3% respondents. Also lack of land and broken
family depicted in Figure-6 caused the migration (Figure-6). It was also observed that
most of these families came from Bhola, a river erosion area of Bangladesh (Table
15).
Figure 6. Causes of migration (%)
100
Causes (%)
90
80
70
60
50
34.9
40 31.8
30
15.5
20 11.6
10 4 2.3
0
Lack of employment Poverty River erosion To pay the loan Others Desire to earn more
money
Table 15. Location of migrated people who came to Dhaka (%)
District Name Percentage

Bhola 61.7
Kishoregonj 9.3
Faridpur 5.6
Barishal 4.7
Shariatpur 4.7
Netrokona 2.8
Others 11.1
Usually these families came to Dhaka and took refuge in the slums. In most of the
cases it was found that CDWs’ fathers were working as rickshaw pullers and mother
working as garment workers (Matrix 1). These families needed more income to lead
their livelihood. It was quite impossible for the families to lead their livelihood with the
income of one earning member. So, these families engaged all of their members in
IGA. The study also observed that mothers put their daughters in domestic work as
she was already involved in such work. On the other hand, in the case of rural areas
relatives and neighbours arranged domestic work for children after observing the
financial crisis of the CDWs families (Box 1).

Box 1.
Laiju (10) used to study at BRAC school and completed her first grade. Her family used to
live in Bhola. Her father borrowed Tk.100,000 to give dowry in his daughter’s wedding. The
whole family escaped to Dhaka after failing to repay the loan. The family lives in a slum in
Basila, Mohammadpur, Dhaka. To lead their livelihood her parents put all of their children in
IGA. Her mother also got involved in domestic work. Sometimes she also went to
employer’s house along with her mother. Gradually she involved herself in domestic work in
the employer’s house, where her mother was working.
Sahana Akter (13) child domestic worker. She completed her second grade. She lost her
father long time ago. After death of her father, the family falls in financial crisis. Then her
mother started begging in a village Chadpur in Matlab. She has seven siblings. By
observing their financial crisis, their neighbour proposed her mother to do domestic work at
Dhaka. Her mother did not agree to go to Dhaka. Because in the absence of mother there
was no one to look after her children. Then her mother decided to send Sahana at Dhaka
for domestic work.
In essence, this stated condition in urban areas compelled parents to involve children
in IGA. We also found that children involvement in IGA were higher in urban areas
(33.5%) compared to rural areas (14.5%) (Fig.7).
Figure 7. Economically active children (5-17 years) in urban and rural area
100 Economically active childrn (%)

90
80
70
60
50 35.5
40
30 14.5
20
10
0
Rural Urban
We also asked the parents in what age they would like to involve their children in IGA.
About 55% of parents in urban areas mentioned that they would engage their
children in IGA before reaching 18 years of age. While 32.5% of the parents in rural
areas mentioned that they would engage their children in work before reaching 18
years of age. It means that parents from urban areas were more interested to put
children in IGA than the parents of rural areas (Fig. 8).

Figure 8. Parents wanted to engage children in IGA below 18 years of age
Parents opinion (%)

100
80
54.6
60
32.5
40
20
0
Rural Urban
The parents put their children in domestic work due to poverty which appears to be a
cause for overwhelming number of cases under consideration. A negligible
percentage of CDWs’ parents in rural areas mentioned other causes such as repay
the loan which had been borrowed to give dowry in daughters’ wedding (3.4%),
fathers’ sickness (1.7%) and father did not look after the family (1.7%). Parents from
urban area also mentioned other causes such as large family size (4.1%), family
needs more income (4.0%), and savings money for CDWs wedding (2.0%) (Fig. 9).
Figure 9. Causes of becoming CDW (%)
Causes (%)
100 93.2 89.8
90
80
70 Rural Urban
60
50
40
30
20
10 4.1 3.4 4 2 1.7 1.7
0 0 0 0 0 0
0
Poverty Large family size Repay dowry Family needs more Save money for Fathers' sickness Father didn't look
income marriage after the children

Along with poverty there were other causes such as broken family, death of father,
and fathers did not look after the family put children in domestic work (Box 2).
Box 2.
Ayesha (15) never went to school. Her father was a rickshaw puller and mother was a
housewife. Her father divorced her mother and re-married in Dhaka. Her step mother did
not look after her. She used to scold her and did not provide her food. Her father’s financial
condition was also not good. Considering the situation, her aunt proposed to her father to
involve her in domestic work. She also agreed to this proposal as she wanted to get rid of
this situation.
Sapna,(10) works in a house as domestic worker. Her family consists of five members, lives
in Tangail. Sapna used to go to BRAC school. She left school while she was in fourth grade.
She is the eldest one in her family. Her father was a gambler and did not earn anything. Her
mother worked as a garment worker in Dhaka. But it was not possible to maintain the family
with her mother’s income. So, the family went back to Tangail. Her mother started to work
in the programme ‘Food for work’. Because of poverty, her parent sent her for working as a
domestic worker to a family living in Dhaka. She was seven years old then. A friend of his
father worked as the mediator. She wanted to continue her education but her family sent
her to work against her will because of poverty.
All these domestic workers were hired to do household tasks. CDWs were hired to
clean houses (29.8%), washing clothes (22.8%) and dishes (18.0%), looking after
employers’ kids (14.0%), cooking (6.1%), cutting vegetables/fish/grinding spices
(4.8%), and shopping (4.4%) (Table 16).
Table 16. Types of households work CDWs had to do (%)
Types of work Percentage

Cleaning house 29.8
Washing clothes 22.8
Washing dishes 18.0
Looking after employers children 14.0
Shopping 4.4
Cooking 6.1
Cutting vegetables/fish/grinding spices 4.8

Areas from where CDWs came to Dhaka (%)

According to the map most of the CDWs came to Dhaka from Mymensingh (36.4%).
There were 30.3% CDWs who came to Dhaka from Bhola, 18.2% from Kishoreganj
and 12.1% from Kurigram. A few CDWs came to Dhaka from Jamalpur (3.1%) and
none from Chapai Nawababganj.
domestic work
We asked CDWs’ parents about the conditions under which they would stop their
children from engaging in domestic work. Parents from both the areas most
frequently responded that if family had sufficient income they would not engage their
children in domestic work. A few percentage of CDWs’ parents from rural areas
mentioned other conditions such as if family could arrange money for CDWs
marriage (5.1%), repay the loan (2.5%), health of earning member improves (2.5%)
and family could ensure security for CDWs (1.3%). Parents from urban areas also
mentioned that if family could arrange money for CDWs marriage (3.8%), they would
not engage their children in domestic work. They also responded other conditions
such as if they could arrange money to build new house in their village (7.7%) and if
their children became adult they would engage them in other professions instead of
domestic work (7.6%). Except one all these conditions would indirectly reflect to the
better financial condition of CDWs’ families in future (Table 17).
Table 17. Condition under which parents would stop their children from
engaging in domestic work (%)
Conditions Rural Urban

(n=50) (n=47)
Family has sufficient income 89.2 73.1
Family can arrange money for CDW’s marriage 5.1 3.8
Family can repay the loan 2.5 -
Health of earning member improves 2.5 7.7
Family can arrange money to build own house - 7.7
When child become adult - 7.6
Family can ensure security for CDW 1.3 -
Multiple responses
CDWs also mentioned that they would not involve themselves in domestic work if
their families financial condition improves so that they could repay loan, buy land, and
build own house in their village. Savings for their better future or wedding would also
stop them from engaging in domestic work. Some of the CDWs mentioned that as
they were aged < 18 years and they were not allowed to work in garment industries.
When they would be 18 years old they would involve themselves in garment
industries instead of working as domestic workers. Because salary in garments
industries was higher than domestic work. A few of the CDWs also mentioned that
they could do nothing to survive except by engaging themselves in domestic work.

Because they were not educated or they did not have skilled training on IGA. One of
our CDWs mentioned that her parents borrowed Tk. 8,000 in advance from her
employer. Her mother assured her that she would bring her home after the money
would be repaid.
Working environments of child domestic workers
About 40% of the CDWs from rural areas received food, lodging, and clothing
without salary from employers. While 85% of the CDWs from urban area received
food, lodging, and clothing with salary from employers. This figure shows that CDWs
from rural areas received less facilities from employers compared to CDWs from
urban area (Fig.10).
Figure 10. Facilities received from employers (%)
Facilities received from employers (%)

100
85.7
90
80
70 59.3
60
50 40.7
Rural
40
Urban
30
20 14.3
10
0
Salary, food, lodging and clothing Food, lodging and clothing without
salary
The findings also show that CDWs from rural areas were less paid than the CDWs
from urban areas. Most of the CDWs in urban and rural areas received Tk. 501-1,000
per month as salary. Although, significant difference was found between urban and
rural areas in this regard, about 17% of CDWs in urban areas was paid Tk.1,001 to
1,500 as salary per month which was higher compared to rural areas. A little over
20% of CDWs in rural and 17.5% of CDWs in urban areas paid <500 taka as salary.
Only 5.3% of CDWs in urban areas and 13% of CDWs in rural areas were paid Tk.
1,501-2,000 per month as salary. Most of the CDWs both in urban and rural areas
paid Tk. 501-1000 as salary (Table 18). Similar results have been found in the
distribution of CDWs salary given to parents (Table 19).

Table 18. Distribution of CDWs by salary (%)
Salary (in Taka) Rural Urban p value

(n=50) (n=47)
<500 26.1 17.5 .517
501-1,000 56.5 60.0 .003
1,001 to 1,500 4.3 17.5 .013
1,501-2,000 13.0 5.0 .807
Table 19. Distribution of CDWs who given salary to their parents

each month (%)
Salary (in Taka) Rural Urban p value

(n=50) (n=47)
<500 23.8 21.1 .216
501-1,000 47.6 63.2 .000
1,001- 1,500 14.3 10.5 .522
1,501-2,000 14.3 5.3 .807
CDWs mentioned that employers assured their parents that they would arrange
marriage for their daughters in future. Employers would bear the cost of CDWs
marriage if they work at employer’s house for long time. Usually CDWs’ parents
directly came to employers’ house to receive the salary of CDWs’ or employers’ sent
salary to the parents. CDWs mentioned that their salary contribute to their parents’
livelihood expenditure such as family could repay the loan and pay for the treatment
of earning member of the family.
A little more than half of the CDWs’ parents from both the areas mentioned that their
children did not mention any problem faced at work place. Those who mentioned
about problems faced at work place, mostly mentioned about too much of work
load, scolded/verbally abused, and beaten by employers. A few parents from both
the areas mentioned about other problems faced at work place such as less food
provided, did not allow children to go home, and low wage (Table 20).
Table 20. Problems faced by CDWs at work place (%)
Problem faced at work place* Rural Urban

(n=50) (n=47)
Too much work load 23.7 22.4
Scolded/verbally abused 20.3 22.4
Beaten by employers 13.6 6.1
Less food provided 8.5 2.0
Did not allow CDW to go home 1.7 2.0
Low wage 3.4 10.2
Did not mentioned any problem 66.1 59.2
*Multiple responses

CDWs also mentioned that the main problem at work place they faced was work
load and irregular payment. They also experienced a bitter condition like beating,
scolding, did not allow CDWs to go home and provided less food. Usually they woke
up 5 to 6 am and went to sleep at 11 to 11:30 pm. They had to clean house, wash
clothes and dishes, look after employers’ children, cook, cut vegetables/fish,/grind
spices, and shopping as a domestic worker. They rarely had a leisure time during the
day. CDWs mentioned that if they did any mistake in work then employers beaten
them and verbally abused them. They also mentioned that employers neglected
them, provided leftover and stale food. A few CDWs mentioned that employers
provided them food as their requirement, shared the same food with employers’, and
have leisure time to watch television at employers’ house (Matrix 1).
CDWs’ parents were asked about the mode of contact by CDWs with their family.
CDWs’ parents in urban areas most frequently responded that they could visit to
CDWs at employers’ house (75%) and CDWs also could visit their family members
(41.7%). While most of the CDWs in rural areas contacted over phone (71.2%). Only
28.8% of CDWs’ parents in rural areas responded that family members had the
opportunity to visit CDWs. No parents in urban areas mentioned that CDWs had not
contacted with parents whereas 1.7% of the parents in rural areas mentioned that
they had no contact with the CDWs. It means CDWs from rural areas got less
opportunity to meet with their family members (Table 21).
Table 21. Mode of contact by CDWs with family (%)
Mode of contact Rural Urban

(n=50) (n=47)
Over phone 71.2 54.2
Family members visit CDW 28.8 75.0
CDW visited family members 10.2 41.7
Through relatives/neighbours 5.1 12.5
No contact with parents 1.7 0.0
Multiple responses
Respondents’ knowledge on child rights
To understand parents’ knowledge on child rights we asked parents- 1) Under what

age an individual was called child, 2) Earning age of an individual and 3) Demerits of
engaging children in domestic work.
According to the United Nations (UN) convention of child rights and based on our
definition of child age we considered an individual as called child up to 18 years of
age. We also considered that an individual become an earning person at 18 years of
age. Among the respondents who could answer the questions according to our
definition, we considered that the respondents had knowledge on child age and
working age. The findings show that a negligible number of respondents both in
urban and rural areas with and without CDWs had correct knowledge about child
age. Although, knowledge on child age with and without CDWs found statistically

significant within rural areas. In contrast, there was no significant difference was
found with and without CDWs in urban areas (Table 22). CDWs also had no
knowledge about the age limit of a child. According to CDWs’ views child refers to a
person who was not able to do any hard work, mothers feed the food, mothers
needed to hold them into laps as well as they were less experienced about the
practical world (Matrix 1).
Table 22. Under what age an individual is called child (%)
Knowledge on child Rural Urban p value

age With CDW Without CDW With CDW Without CDW 1vs 2 3 vs 4
(n=50) (n=1,404) (n=47) (n= 63)
1 2 3 4
Had knowledge on 8 1.6 4.3 6.3 .001 .636
child age (up to 18
years)
Did not know 92 98.4 95.7 93.7 .001 .636
Almost similar findings were observed about the knowledge on earning age of an
individual. But this knowledge between urban and rural areas with and without CDWs
was found statistically insignificant (Table 23). CDWs also had no knowledge in this
regard. According to their views when a person could look after himself/herself, had
exposure with practical life then she/he would become fable to earn money (Matrix 1).
Table 23. Under what age an individual becomes earning person (%)
Knowledge on the Rural Urban p value

age of earning With CDW Without CDW With CDW Without CDW 1vs 2 3 vs 4
person (n=50) (n=1,404) (n=47) (n=63)
1 2 3 4
Had knowledge on 14 14.7 6.4 12.7 .895 .279
earning age of a
person (18 years)
Did not know 86 85.3 93.6 87.3 .895 .279
Most of the parents in both the areas responded that children could be sick due to
heavy work load, get injured at work, education could get affected, and they could be
deprived of familial affection for staying at a distant work place. A moderate
percentage of parents from the both areas cited that children could be physically,
verbally and sexually abused. A small percentage from both areas parents of CDWs
and without CDWs mentioned that they did not know about the demerits of engaging
children in domestic work (Table 24).

Table 24. Demerits of engaging children in domestic work (%)
Demerits Rural Urban

With CDW Without With CDW Without CDW
(n=50) CDW (n=47) (n=63)
(n=1,404)
Become sick due to work load 76.6 71.7 57.5 65.1
Children education could get 68.0 49.8 29.8 33.3
affected
Get injured by doing domestic work 64.0 49.8 34.0 36.5
Deprived from familial affection 50.0 38.1 53.2 55.6
Physically abused 24.0 14.7 36.1 33.7
Sexually abused 2.0 .8 10.6 11.1
Verbally abused 2.0 2.2 8.5 3.2
Do not provide food as required - - 12.8 4.8
Chances of being killed by 4.0 .8 - 9.5
employers
Future is ruined - - 2.1 -
Less/Irregular payment 2.0 .5 - -
Difficult to arrange marriage for girls .0 1.7 - -
Lack of freedom .0 .2 - -
Did not know 4.0 2.6 8.5 3.2
Multiple responses

Matrix 1. Case studies of CDWs
Characteristics Description (number)

CDWs Fathers’ occupation Rickshaw puller (4), Day labour (non-agri) (3),
unemployed (1) fathers not alive (1).
CDWs Mothers’ occupation Domestic worker(7), housewife (2) and beggar (1)
Education level of CDW No education (6), primary education (3), religious
education (1).
Age range of CDWs <10 (1), 10-12 years (4), 13-15 years(5),16-17
years(1)
Sex Female
Causes of migrate to Dhaka Poverty, river erosion, lack of land, death of father
and break up family
Person who put children in domestic Parents, relatives, neighbour
work
Causes of getting involved in Poverty, death of father , father didn’t look after the
domestic work family and broken family
Remuneration of CDWs Food, lodging and clothing without salary (1)

<Tk.1,000 (8)
Tk.1,001-1,500 (1)
Tk.1,501-2,000 (1)
Working environment
Non-congenial environment Too much work load, scolding, provided less,
stale and leftover food and beating.
Congenial environment CDWs were provided required food; share the
same food with employers, have leisure time and
could watch television.
Types of work Cleaning house, washing clothes and dishes,
looking after employers’ kids, cooking, cutting
vegetables/fish/grinding spices and shopping
CDWs knowledge on child rights
Under what age an individual is <10 years (6)
called child 11-12 years (2)
13-14 years (2)
Under what age an individual <10 years (3)
becomes earning person 11-12 years (2)
13-14 years (2)
15-16 years (4)
Figure in parentheses indicates the no. of observation

4. Discussion and conclusion
The study aimed to identify households with and without CDWs to create a data
bank consisting demographic characteristics. The study also aimed to explore the
cause and process of becoming CDWs in Bangladesh.
We have been found that the demographic and socioeconomic characteristics of

households with and without CDWs were almost same both in urban and rural areas.
So, the households without CDWs have the possibility to involve their children in any
IGA including domestic work. Because poverty is the root cause to involve children in
domestic work. Parents in urban areas and CDWs, themselves were the most
instrumental in getting children engaged in domestic work. In contrast, relatives,
neighbours, middleman, and employers in rural areas was most influential to get
children engaged in domestic work.
The study observed that poverty was the main cause to involve children in domestic
work and all these workers came from poor and extreme poor families which are
similar to the findings of other studies (Blanchet 1996; Pelto 1997 and Rahman
1995). We observed that the national average annual per household income and
expenditure were much higher than those of the surveyed population (BBS 2012).
We also observed a good proportion of households with and without CDWs had no
land, education or savings. While, access to land, education, savings and income
can ensure to meet any crisis (Islam 2005). In any kind of crisis situation of the family
like reduction of family income, failure of crops, natural disaster, and high
unemployment or recession the poor and extreme poor families usually involves their
children in labour force including domestic work (Bourdillon et al. 2010). So,
households’ socioeconomic status compelled parents to involve their children in
labour force including domestic work.
Majority of the respondents in study area relies on formal and informal sources of
treatment (e.g., drug shop, govt. hospital) which is a common practice for the poor
and marginalized population in Bangladesh. Financial barriers prevent them to seek
treatment from formal sector (Ahmed 2011). Both qualitative and quantitative analysis
under the study reveals that sickness of earning member in the family compels
children to work to minimize the cost of living especially during ailment of the
incumbent income earner.
The study found that in urban areas parents and CDW herself/himself were the most
instrumental to get into in domestic work. Like other study, we also found that rural
poor people migrate rural to urban areas due to poverty, lack of job opportunity, river
erosion, rupture of family relationship, and scarcity of land (Zohir 2001). In relation to
survive in a city they involved their children in informal labour sector, like domestic
work (Farhana, Marchi and Rahman 2010). The study observed that in urban areas

children’s mothers were working as domestic worker. They often bring their children
at work place along with them. Subsequently, these children follow the path of their
mothers which ended up to be employed as domestic worker (Pelto 1997). To cope
with the poverty in urban slum in Dhaka city, parents spontaneously involve their
children in labour force including domestic work.
In rural areas, social network of the families such as relatives, neighbours, middleman
and employers influences children to get engaged in domestic work (Save the
children 2009, Gianni 2006 and Pelto 1997). Middlemen also plays a vital role to find
domestic work for children especially in rural areas (Dostie and Vencatachellum
2004). So, community has the greater influence to get children engaged in domestic
work.
In conclusion, we found that the demographic and socioeconomic characteristics of

surveyed population were strongly associated with poverty. Households with and
without CDWs socioeconomic and demographic characteristics were almost same.
Poverty was the root cause for involvement of children in domestic work. Parents,
neighbours, relatives, and CDWs, himself/herself were the most influential person to
engage children in domestic work. The study also created a data bank consisting the
demographic characteristics of households with and without CDWs. We speculate
that the data bank would be able to help BRAC HRLS programme to take an initiative
by targeting poor and extreme poor households to engagement of their child in
domestic work. In other words to safe save the child rights acclaimed by all
concerned.

5. Recommendations
The findings suggest that child domestic workers suffer from lack of affection, leisure,
education, health support and their rights to live with their families. Therefore, this
research aims to find out how children can live with their families and communities as
much as possible. Nevertheless, poverty and other social factors may not allow them
to have this opportunity. The study also found that communities including parents
were the instrumental to involve children in domestic work. In any civilized society this
system of child labour is not acceptable. Therefore, GO, NGO like BRAC can take
programmes or initiative to reduce children’s involvement in domestic work which is
at times risky as inhuman. The study created a data bank consisting demographic
profile of households with and without CDWs with location. By using this data bank,
BRAC can extend its services, i.e., economic, health, education to improve the
livelihoods of poor and extreme poor people and create awareness to prevent
children involvement in domestic work at community level in Bangladesh.
Furthermore, we also recommend that a qualitative study can be conducted to
explore the social factors more in-depth that compel children to engage in domestic
work.

References
Ahmed SM, Hossain S, Khan AM, Islam QS and Kamruzzaman M (2011). Lives and livelihoods
on the streets of Dhaka City: Findings from a population based exploratory survey. Dhaka:
BRAC. (RED Working paper No. 19).
Bangladesh Bureau of Statistics (BBS) (2012). Preliminary Report on Household Income &
Expenditure Survey-2010. Dhaka: Bangladesh Bureau of Statistics.
Bangladesh Bureau of statistics (BBS) (2011). Population and Housing Census. Preliminary
results.
BBS and UNICEF (2006). Baseline survey on child domestic worker. International labour
organization . Dhaka.
Bangladesh Bureau of Statistics (BBS) (2007). Household income and expenditure survey-
2005. Dhaka: Bangladesh Bureau of Statistics.
Bangladesh Bureau of Statistics (BBS) (2005). Baseline survey for determining hazardous child
labour sectors in Bangladesh.
Bangladesh Bureau of statistics (BBS) (2003). Child labour survey in Bangladesh. Ministry of
planning, Government of the people’s Republic of Bangladesh.
Black M (1996). A hand book on child domestic workers: findings a voice. Sussex: Anti-slavery
international.
Blanchet T (1996). Lost innocence stolen childhoods. Dhaka: The University Press.
Bourdillon MFC, Levison D, William EM and White B (2010). Rights and Wrongs of children’s
work. New Brunswick, NJ University press Rutgers.
BRAC (2009). Pathways out of extreme poverty: findings from round I survey of CFPR phase II.
Dhaka: BRAC, IV, 264P.
Burra N (1997). Born to work: child labour in India, Delhi: Oxford University Press.
Dixit A (1997). Labour force participation of girl child in India, The Indian Journal of Labour
Ecnomics, 40(4).
Dostie B and Vencatachellum D (2004). Compulsory and voluntary remittances:

Evidence from child domestic workers in Tunisia. HEC Montreal, Universite de Montreal,
Quebee, Canada.
Elgbeleye OS and Olasupo MO (2011). Parental Socio-Economic status as correlate of

child labour in Ile-Ife, Nigeria. Educational Research and Reviews 7(8);201-6.

Farhana MK, Marchi DR and Rahman SA (2010). Surviving pattern of women and children
migrants in urban Bangladesh. Bangladesh e-Journal of Sociology 7(2): 50-64.
Gianni L (2006). Migration and education : child migrants in Bangladesh. University of Sussex,
Sussex centre for Migration Research. (Sussex migration paper no.33)
Hawamdeh H, Spencer N (2003). Effect of work related variables on growth among

working boys in Jordan. J Epidemiol Community Health 57:154–
158.doi:10.1136/jech.57.2.154
ILO child labour statistics. available at http://www.ilo.org. (Accessed on 20.05.2012)
Islam N (2005). Dhaka New: Contemporary urban development. Dhaka: BGS publication.
Kielland A and Tovo M (2006). Children at work child labour practices in Africa. Boulder, CO
Lynne Rienner.
Pelto B (1997). Daily lives of working children: Case studies from Bangladesh. UNICEF
Bangladesh.
Rahman H (1995). Child domestic workers: Is servitude the only option ? Dhaka: Shoishab
Bangladesh.
Save the children (2009) Living inside room outside law; A study on child domestic worker and
the role of government and civil society. Dhaka.
Zohir CS (2001). Social impact of the growth of garment industry in Bangladesh. The quarterly
Journal of the Bangladesh institute of development of studies XXVII (4):41-80.

Annexure
Annexure 1. Sampling framework
Sampling
Six districts purposively selected from four division of Bangladesh

(Dhaka, Rangpur, Barisal, Rajshahi)
(Considered development indicators, i.e., property ratio, literacy rate,

school attendance rate of children aged .5years and above, and
disaster prone areas)
Six districts purposively selected from each district

(Mymensingh, Kishoreganj, Jamalpur, Kurigram, Chapai
Nawabganj, Bhola)
(Considered by BRAC’s programme activities, i.e. BEP, ADP, BHP,

TUP, HRLS, CEP)
Three villages selected randomly from six upazilas
(Considered to be least developed and poverty prone areas)

Total 18 villages
110 slum dwellers selected from 1,454 poor and extremely poor
one slum in Dhaka for interview households selected from 18 villages for
interview
(Considered TUP programme of
BRAC)
(At least 60 HH from each village)

Annexure 2. Conduct social map and wealth ranking
(Annexure 2 continued...)

(...continued Annexure 2)

Annexure 3. Livelihood resources and wealth ranking
The wealth ranking provides a picture of different wealth groups that exist in the
sample study areas. In accordance with the pattern of resources, the households or
villagers identified four wealth groups in the 18 study villages which were located in
six districts of Bangladesh. People require a range of resources to secure positive
livelihood outcomes. The evaluation tried to understand the strengths of resources of
the study area people and then selected real target people of the CDW study.
Available livelihood resources of four identified wealth groups were presented in
matrix 2.
Matrix 2. Wealth category and livelihood resources in study areas
Wealth Livelihood resources /indicators Remarks

category
Rich • Possess 250 to 800 decimals of land. Not included in
• Surplus food over the year. CDW study
• Monthly income range Tk. 30000-50000
• Stable sources of income from service, agriculture business
and shop.
• Possess tin roofed house or building tubewell, sanitary latrine,
pond, livestock and poultry.
• Sons do work in foreign countries.
Middle • Possess around 50 to 249 decimals of land. Do
• Possess shops, small business and low level job.
• Monthly income range, Tk. 11000-30000.
• Do agricultural own land and share cropper, poultry and
livestock rearing, homestead gardening, and rickshaw/van
owner.
• Household size is small.
• Can manage food all over the year
Poor • Own homestead with a few amount of land, e.g., 10-49 CDW study
decimals targeted
• Monthly income range, Tk. 5000-10000. household
• Sell manual labour (agri. or non-agri.)
• Do rickshaw/van pulling and fishing as a fisherman
• Possess small business, poultry and livestock.
• Produce vegetables in leased land.
• Big household size with young children and they worked like
maid servant
• Frequently migrate out for employment.
• Food deficit for two or three months over the year.
Extreme • Do not own homestead and live on government’s land and Considered
poor embankment. Also some household has a piece of land which CDW study
has less than ten decimals. targeted
• Monthly income less than Tk. 5000. household
• Few are female-headed households with young children and
they worked as a maid servant
• Sell manual labour (agri. and non-agri.), van/rickshaw pulling.
• Produce vegetables in leased land.
• Do seasonally available work and fishing.
• Frequent borrow from others.
• Food crisis over the year.

Annexure 4. Analysis plan
Household
Rural Household Urban Household

(1,454) (110)

Household Household Household Household
(50) (1,404) (47) (63)
Issues the study dealt with:

Household level : Socio-demographic profile, economic empowerment, health status and
health services, association with BRAC and other NGOs.
CDW : Process of becoming CDWs, working environment and remuneration of
CDWs.

Exploring The Causes and Process of Becoming Child Domestic Worker

Uploaded by

Document Informationclick to expand document information

Copyright:

Available Formats

Exploring The Causes and Process of Becoming Child Domestic Worker

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Exploring The Causes and Process of Becoming Child Domestic Worker

Uploaded by

Copyright:

Available Formats

Working Paper No.

Exploring the Causes and Process of

Shuburna Chodhuary, Md. Akramul Islam, Jesmin Akter

Copyright © 2013 BRAC

Page makeup and composition

Design and layout

Research and Evaluation Division

BRAC/RED publishes research reports, scientific papers, monographs, working

Printed by BRAC Printers at Tongi, Gazipur, Bangladesh.

The causes and process of becoming child domestic worker iii

ADP Adolescents Development Programme

iv RED Working Paper No. 35

The causes and process of becoming child domestic worker v

The study aimed to explore the demographic and socioeconomic characteristics of

vi RED Working Paper No. 35

Children’s involvement in domestic work for income is a common phenomenon in a

The causes and process of becoming child domestic worker vii

Four economic empowerment indicators, e.g., income, expenditure, savings, and

Health status and health services

Becoming child domestic workers

viii RED Working Paper No. 35

Most of the respondents came to Dhaka to seek employment (34.9%), poverty

The causes and process of becoming child domestic worker ix

Working environments of child domestic workers

Respondents’ knowledge on child rights

x RED Working Paper No. 35

The causes and process of becoming child domestic worker xi

Children’s involvement in domestic work is an invisible and cheap form of

The causes and process of becoming child domestic worker 1

Rational of the study

However, culturally children engagement in domestic work is accepted and practiced

2 RED Working Paper No. 35

• Create a data bank containing demographic information on parents, families and

The causes and process of becoming child domestic worker 3

Study area and population

Table 1. Study site

Division District Upazila/Area Number of Number of

4 RED Working Paper No. 35

Methods of data collection

Data management and analysis

The causes and process of becoming child domestic worker 5

6 RED Working Paper No. 35

In considering the first objective of the study the socio-demographic characteristics

The causes and process of becoming child domestic worker 7

Indicators Rural Urban Difference p value

Economic empowerment indicators

Four economic empowerment indicators, e.g., income, expenditure, savings and

8 RED Working Paper No. 35

Table 3. Distribution of household by income

Annual income Rural Urban p value

Figure 1. Major sources of household income in a year

Major sources (%)

The causes and process of becoming child domestic worker 9

Expenditure and major sources of household expenditure

Household expenditures were estimated based on the average expenditure of

Table 4. Distribution of household by expenditure

Annual expenditure Rural Urban p value

10 RED Working Paper No. 35