Impact of Interventions on Child Malnutrition and Youth Unemployment in Eastern Congo Skyler Barlow . David Hobson . Brandon Pichanick . Diego Pinzon Cait Stadler . Julie Swensen . Gabriel Vander Hey Faculty Advisor: Dr. Kishore Gawande
Dec 02, 2014
Impact of Interventions on Child Malnutrition and Youth Unemployment in
Eastern CongoSkyler Barlow . David Hobson . Brandon Pichanick . Diego Pinzon
Cait Stadler . Julie Swensen . Gabriel Vander Hey
Faculty Advisor: Dr. Kishore Gawande
Roadmap• Background• Partners and Pre-Survey• Project 1: CEFADES• Project 2: Nutrition• Conclusion• Questions
The Democratic Republic of the Congo
Conflict in North Kivu• Crossroads for armed groups
• Butembo population
• Stability of Butembo
Client & Partner Background• Client: The Conflict & Development Lab• Partner: The Universite Catholique du Graben
– Students will be co-authors– CEFADES & Giorgio Cerruto Therapeutic Nutrition Center
Pre-Survey• Gavin Finnegan
• Purpose– Train enumerators– Identify potential
weaknesses of the surveys
Project I: CEFADESYouth Employment
What is CEFADES?
• Youth Vocational Program• Focused on the Disenfranchised• 594 since 2009
– 30% female• Skill Training
– mechanics, carpentry, masonry, dressmaking, hairdressing, culinary arts, etc...
CEFADES Research Questions
•Does employment increase?
•Does income increase?
CEFADES Survey
CEFADES Methodology• Comparison of Control and Treatment• OLS regression on income• Comparison of Control and Treatment• Probit regression on working status• Child soldier regressions
Employment ImpactThe Proportion of Populations Working vs Unemployed
Employment ImpactDependent Variable: Working Status
Income ImpactAverage Monthly Income in Dollars
Income ImpactDependent Variable: Log of Monthly Income
Dependent Variable: Log of Monthly Income
Bottom Line• Research Questions
– Does employment increase?Yes
– Does income increase?Yes
• Limitations– Post-Treatment Survey
Ideal: Panel Data– Non-Randomized Treatment
Project II: Nutrition
Setting the Stage
Source: Eastern Congo Initiative
Nutrition Research Questions
• What is the prevalence of child malnutrition in the Bunyuka Parish?
• In comparison to the control, does the treatment positively affect the nutritional status of children?
Nutrition Methodology
• Control Study– WHO Standardized Measurements
• Average Treatment on the Treated (ATT)– Independent Variable: Nutritional Therapy– Dependent Variables: Height-for-Age (HFA), Weight-for-Age (WFA)
Nutrition Control Survey
Nutrition Treatment Survey
Nutrition Sample
Untreated Group• 714 households surveyed• 1,750 children measured
Treatment Group• 215 households surveyed• 351 children measured
io
Prevalence of MalnutritionControl Sample
Comparison of Below Average WHO Outcomes w/ Above Average Bunyuka Outcomes
Selection on Observables• Self-Reported Nutritional Status• Education Levels• Distance to Health Center• Nutritional Preferences• Age • Ethnicity/Language• Cost of Treatment
Quartile Treatment Effect
Average Treatment Effect
Bottom Line
• Research QuestionsWhat is the prevalence of child malnutrition in Bunyuka Parish?
Very prevalentIn comparison to the control, does the treatment positively affect the nutritional status of children?
Yes, but it is best for children under 3 years• Limitations
–Not a perfect control–Post-treatment survey only
CONCLUSION
Conclusion
CEFADES•Positively impacts youth•Confirms training effectiveness•Contradicts child soldier literature•Unique program for this part of the world
Conclusion
Nutrition• Context• Survey Mechanics• Findings–Rural health outcomes–Observable costs to treatment–Ideal age range for treatment
Conclusion
Client•Program evaluation capabilities
Field of Conflict & Development•Research capabilities
Questions
Skyler Barlow | David HobsonBrandon Pichanick | Diego Pinzon
Cait Stadler | Julie Swensen | Gabriel Vander Hey
CEFADES Selection Process
1 - Program Psychologist
2 - NGO’s and Government
3 - Parents
4 - Advertisement in Churches
Survey
Key Factors Measured
1 - Demographics
2 - Income
3 - Health
4 - Spillover Effects
Survey: By Gender Comparison
CEFADES - PSM results