ANALYSIS OF PROTECTIVE PATHWAYS IN AMHARA REGION
MAJA GAVRILOVIC @SAVAGEWITHIN & ELSA VALLI @VALLIELSA
CHILD MARRIAGE & ETHIOPIA’S PRODUCTIVE
SAFETY NET PROGRAMME

01.
BACKGROUND & RATIONALE
02.
CONCEPTUAL FRAMEWORK & METHODOLOGY
03.
RESULTS
04.
IMPLICATIONS FOR POLICY & PROGRAMMING
OVERVIEW

DESPITE PROGRESS, MANY GIRLS
REMAIN AT RISK OF CHILD MARRIAGE
Ethiopia: Child marriage (CM) prevalent but declining
Prevalence has dropped by 20 percentage points in last decade
2005: 59% of young women married before 18
2015: 40% of young women married before 18
Amhara region has lowest median age at marriage: 15.7 years
Rate of decline insufficient to eliminate CM by 2025

Cash transfers a promising strategy to reduce
CM at scale
PSNP delayed out-migration of adolescent girls
Girls act as substitutes for adult labour 
potential spillover effects on delaying CM
Evidence remains limited & mixed
Aim: Explore pathways through which PSNP
may impact CM
Evidence will inform PSNP5, intentional links to
CM programming
WHY WE DID THE STUDY

Descriptive mixed-method study explores CM
dynamics and pathways of change
Quantitative
Surveys with one female caregiver in 5,355 HHs
Qualitative
15 matched, in-depth interviews with heads of
HH & female adolescents (unmarried & married)
Key Informant Interviews with woreda
representatives
STUDY DESIGN & METHODOLOGY
Does not conclude causal
relationships
Does not make conclusions
for full socioeconomic
distribution of population
Aim Explore how PSNP could
impact decisions on CM &
through which pathways

SETTING
THE SCENE:
CHILD
MARRIAGE
DYNAMICS
IN AMHARA
1. PREVALENCE OF CM
2. ATTITUDES & PERCEPTIONS
3. KNOWLEDGE OF CM LAWS
4. MARRIAGE DECISION-
MAKING DYNAMICS

1. Prevalence of CM in
study sample
Child marriage prevalence by age group (Individual-level)
Differences between ISNP survey and DHS:
• DHS all Amhara – ISNP only 4 woredas
• Different populations (includes large PDS population)
• Lower rates may be result of PSNP effect
Females married as child
(currently residing in the hh)
Females married as child
(DHS)
Age at time of Baseline Survey N Percentage N Percentage
(1) (2) (5) (6)
Females 20-24 466 0.322 277 0.430
Females 20-29 982 0.414 594 0.531
Females 25-29 516 0.498 317 0.617

ATTITUDES & PERCEPTIONS:
16.6 YEARS IDEAL AGE FOR MARRIAGE
All HHs
(Mean)
HHs with females 20-
24 years old
(Mean)
Ideal age a girl should get married 16.60 16.73
Ideal age a girl should get married: U18 0.60 0.56
Ideal age a girl should get married: U15 0.08 0.04
Age at which adolescent females usually get married in this community 16.63 16.77
In this community females usually get married before 18 years 0.60 0.57
In this community females usually get married before 15 years 0.10 0.10
N 5,355 536

ATTITUDES & PERCEPTIONS:
PHYSICAL MATURITY IS MAIN INDICATOR OF ‘BEING
READY’ FOR MARRIAGE
Everyone marries at this age 0.26
Will have finished desired schooling 0.16
Will be physically matured enough 0.56
Will be mentally matured enough 0.31
Will be time to start having children 0.19
It is the legal age 0.06
To avoid pre-marital sexual relationships 0.06
N 5,361
Girls generally described as
ready for marriage once show
physical signs of puberty.
“She [is ready] when she starts
seeing menstruation & she has
the ability to become pregnant.”
- unmarried girl, Libo Kemkem

ATTITUDES & PERCEPTIONS:
DRIVERS COMPLEX, MULTIFACETED & ENTANGLED
Poverty one among many factors…
Gender & religious norms restrict girls’ premarital
sexual activity, maintain chastity & family honour
Desire to consolidate or demonstrate wealth through
marriage & expand social ties with better-off families
Marital norms influence parents’ decisions & limit
agency to avoid or break marriage proposals
Limited value placed on girl’s education exacerbated
by dearth of alternative options for girls

“Our culture also forces us to
let her get married at that age
[15] because a girl might be
exposed to unnecessary
relationship with a male.”
- parent of unmarried girl, Dewa Chefa

“People say prostitute & tramp
if a girl gets married at a later
age. People say nothing if a boy
gets married at a later age.”
- married girl, Dewa Chefa

KNOWLEDGE OF CHILD MARRIAGE LAWS IS LIMITED
“The ideal age to get
married for girls is 15
years old. […] Below this,
now days it is illegal.”
- unmarried girl, Libo
Kemkem
N All
Knows there is a legal marital age for boys & girls 5,362 0.08
Knows legal marital age for girls
(among those that know there is a legal age)
1,371 0.80
Knows legal marital age for girls
(among all respondents)
5,362 0.21
Knows marital legal age for boys 439 0.32
ONLY 20% OF RESPONDENTS REPORT CORRECT LEGAL AGE

Father Mother Parents
Person
getting
married
Parents with
person getting
married
Who decides at what age a female
should get married
7.3 4.4 58.7 18.1 11.4
Who should decide at what age a
female should get married
5.8 1.3 60.4 19.1 13.3
Who decides who a female marries 6.2 3.0 52.5 22.3 15.8
Who should decide who a girl marries 5.1 0.8 52.1 24.0 17.7
MARRIAGE DECISION MAKING DYNAMICS:
70% OF RESPONDENTS BELIEVE THAT GIRL DOES NOT HAVE A SAY
“Parents usually decide when a girl is ready to get married.”
- unmarried girl, Libo Kemkem

TESTING
POTENTIAL
PATHWAYS
1. IMPROVED ECONOMIC
SECURITY
2. EDUCATION
3. IMPROVED KNOWLEDGE &
ATTITUDINAL SHIFTS

IMPROVED ECONOMIC SECURITY:
DECREASE RISK OF CHILD MARRIAGE
Payments decrease financial burdens & improve economic security
• Reduced economic pressures to marry girls
• Improved ability to keep girls in school: education has multiplier
effects on girls
Some evidence PSNP affects work allocation within HH
• Girls required to help when adults engaged in public works &/or
substitute for adult labour at public work sites
• Negative effects on schooling
Transfer amount based on HH size
• HHs may retain girls to get larger transfers but CM attitudes may
remain unchanged
“If I want to work
on PSNP she
[daughter]
performs
household
work.”
- parent of
unmarried girl,
Dewa Chefa

IMPROVED ECONOMIC SECURITY:
INCREASE RISK OF CHILD MARRIAGE
Additional income may increase HH resources
needed to cover costs of wedding
• Bride price and dowry remain important customs
Participation in the PSNP can increase girls’
prospects for receiving marriage proposals
Socio-economic gains through wealth consolidation
obtained through marriage may outweigh benefits
of PSNP
“Both parents from the bride
and groom side present dowry
for each other. Both families
give money, farmland, &
different materials to boy & girl
who get married to set up their
house. The boy family gives
money to her family for
betrothal (Macha).”
- unmarried girl, Dewa Chefa

EDUCATION PATHWAY
Knowledge & information empowers girls to express & exercise their choices
“We discuss about early marriage at school in girls’ clubs. I discuss with my best friend not to
practice early dating.” - unmarried girl, Libo Kemkem
Access to social networks & support builds girls’ confidence to resist marriage
“Last year one family planned to marry their daughter at the age of 16 years & then she
reported it to her teacher. The teachers intervened with the police & the marriage was
interrupted.” - parent of unmarried girl, Libo Kemkem
Increased awareness of penalties & in-school mechanisms to report child marriage
“In schools if a girl charges her parents for the act of early marriage they can even be jailed
for two or more weeks.” - unmarried girl, Libo Kemkem

ATTITUDINAL SHIFTS AT COMMUNITY LEVEL
Additional components of PSNP4 behaviour change
communication (BCC), case management by social
workers, monitoring of co-responsibilities
BCC key mechanism to shift beliefs & gender norms
related to girls’ value in society & importance of education,
especially if parents & community members involved
Community conversations on harmful traditional
practices tackle community attitudes around CM
Interaction with SWs/Community Service Workers & co-
responsibilities influence awareness & attitudes towards
education & CM

“Because of what we learn at every PSNP
meeting about education, more people are
[planning] to teach their children rather
than pushing for marriage. We are now
focusing more on education.”
- parent of unmarried girl, Libo Kemkem

“The participation of my father in the
PSNP changed views on gender equality.
He respects my mother & discusses with
her to ensure all children attend school
regularly. In our family there is no
problem in sending children to school.”
- unmarried girl, Libo Kemkem

“[Service workers] teach us & we are
happy with their advice. They tell us about
birth spacing. If you give birth frequently,
you may even die. If a girl is exposed to
child marriage & she gives birth it is bad
for both.”
- parent of unmarried girl, Libo Kemkem

CONCLUSIONS
Despite progress, rates of CM
still high in Ethiopia
Interest in understanding effect of
national programmes (PSNP) on
CM
Attitudes towards CM
• Broadly supportive of CM
• Religious & cultural values
perpetuate CM
• CM reinforced by limited voice
& agency exercised by girls

CONCLUSIONS
Economic Pathway
• PSNP reduce risk of CM: reduced financial burden, keep
girls in school, relocate labour within HH
• PSNP increase risk of CM: more resources for bride price
Education Pathway
Courses reduce CM: social networks & support,
empowerment, awareness of penalties, report mechanisms
Attitudinal Shifts Pathway
• BCC & SWs influence decisions & investments of PSNP in
education
• If whole community is engaged, BCC in influence greater

RECOMMENDATIONS & LEARNINGS
To inform PSNP5
1. Additional support for poorest families to retain girls at school
2. Link girls at risk & married/divorced minors to empowerment
programmes
3. BCC module on CM messaging
4. BCC to include all members of HH
5. Strengthen capacity of SWs & HEWs: case management &
referrals to social services

RECOMMENDATIONS & LEARNINGS
To inform programming on CM:
1. Improve community awareness on CM laws
2. Promote dialogue & access to information on
CM in schools
3. Target all community members including
PSNP clients in conversations on CM
4. Develop specific interventions for out-
of/school/married girls, and IDPs

RESEARCH TEAM
UNICEF Office of Research
Maja Gavrilovic, Tia Palermo, Elsa Valli,
Francesca Viola
BDS Center for Development Research
Teketel Abebe, Getinet Mesay Kebede,
Alene Matsentu, Fekadu Muluye, Feredu
Nega, Tadele, Yenenesh Tadesse
UNICEF Ethiopia
Ellen Alem, Getachew Berhanu, Karin
Heissler, Mathilde Renault, Lisa-Marie
Ouedraogo-Wasi, Vincenzo Vinci

• Behrman JA, Peterman A, Palermo T. Does keeping adolescent girls in school protect against sexual violence?
Quasi-experimental evidence from East and Southern Africa. Journal of Adolescent Health. 2017;60(2):184-90.
• Dake F, Natali L, Angeles G, De Hoop J, Handa S, Peterman A. Income transfers, early marriage and fertility in
Malawi and Zambia. Studies in family planning. 2018;in press.
• Handa S, Halpern CT, Pettifor A, Thirumurthy H. The government of Kenya's cash transfer program reduces the
risk of sexual debut among young people age 15-25. PLoS One. 2014;9(1):e85473-e.
• Handa S, Peterman A, Huang C, Halpern CT, Pettifor A, Thirumurthy H. Impact of the Kenya Cash Transfer for
Orphans and Vulnerable Children on Early Pregnancy and Marriage of Adolescent Girls. Social Science &
Medicine. 2015;141:36-45.
• Heinrich CJ, Hoddinott J, Samson M. Reducing adolescent risky behaviors in a high-risk context: the effects of
unconditional cash transfers in South Africa. Economic Development and Cultural Change. 2017;65(4):619-52.
• Hoddinott JF, Mekasha TJ. Social protection, household size and its determinants: Evidence from Ethiopia.
International Food Policy Research Institute (IFPRI); 2017.
• Hallfors D, Cho H, Rusakaniko S, Iritani B, Mapfumo J, Halpern C. Supporting adolescent orphan girls to stay in
school as HIV risk prevention: evidence from a randomized controlled trial in Zimbabwe. American journal of
public health. 2011;101(6):1082-8.
• Mekonnen B, Aspen H, editors. Early Marriage and the Campaign Against it in Ethiopia. 16th International
Conference of Ethiopian Studies; 2009; Trondheim.
• Presler-Marshall E, Lyytikainen M, Jones N, Montes A, Pereznieto P, Tefera B. Child marriage in Ethiopia: A
review of the evidence and an analysis of the prevalence of child marriage
REFERENCES

Maja Gavrilovic
mgavrilovic@unicef.org
Elsa Valli
evalli@unicef.org
@UNICEFInnocenti
@TransferProjct
transfer.cpc.unc.edu
THANK YOU.