Impact Evaluation of a Teenage Mother Program in Colombia

Last registered on April 21, 2017

View Trial History

Pre-Trial

Trial Information

General Information

Title

Impact Evaluation of a Teenage Mother Program in Colombia

RCT ID

AEARCTR-0002122

Initial registration date

April 21, 2017

Initial registration date is when the trial was registered.

It corresponds to when the registration was submitted to the Registry to be reviewed for publication.

First published

April 21, 2017, 2:15 PM EDT

First published corresponds to when the trial was first made public on the Registry after being reviewed.

Locations

Country

Colombia

Region

Latin America

Primary Investigator

Name

Bibiana Taboada

Affiliation

Inter-American Development Bank

Contact Primary Investigator

Other Primary Investigator(s)

PI Name

Arturo Harker

PI Affiliation

Universidad de los Andes

Contact Investigator

Additional Trial Information

Status

On going

Start date

2015-11-02

End date

2019-06-28

Keywords

Education, Health, Labor

Additional Keywords

Teenage pregnancy

JEL code(s)

I12, J13, J24

Secondary IDs

Abstract

The objective of the evaluation is to assess the impact of the "Teenage Mother Program" designed and implemented by Fundación Juanfe (www.juanfe.org), a non-profit and non-gubermental organization working to improve the quality of life of pregnant teenagers and their offspring. The Program is expected to have a positive impact on the mothers' socio-emotional development (primarily on self-esteem, self-efficacy and aspirations), educational attainment (through vocational training), their employability, their access to sexual and reproductive health services, the quality of their relationship with their children, and their access to nutritional and medical assistance for their children. After almost 8 years of in the field experience, the Organization has identified several sources of vulnerability faced by their beneficiaries: extreme poverty (approximately 80% como from families where the average per capita income is USD $ 0.73 a day and income flows are erratic and mostly come from informal activities); sexual abuse (approximately 30% of the beneficiaries have been victims of sexual violence at some point in their lives); gender-based violence (all beneficiaries have been victims of various forms of gender-based violence); and low human capital (frequently girls end up dropping out of school to avoid shame and stigma of being pregnant). Rather than blaming the teenager, the Program targets the risk factors that led to early life pregnancy in the first place: lack of education, poverty, emotional instability, and gender discrimination, among others.

The impact evaluation study is based on a randomized controlled trial (RCT) design at the individual level. The study was designed and is being implemented by the Inter-American Development Bank (IADB) and Universidad de los Andes. The experimental sample for the ongoing RCT has been constructed using an excess demand methodology. Given that the Program's selection process is based on an open call where applicants have to go through three different filters (cognitive tests, emotional tests and a face to face interview), to avoid salient selection biases in our impact estimates the control and treatment groups were constructed from the universe of individuals that were deemed as eligible. Therefore, the experimental sample resulted from the excess demand generated during the admissions process of the two cohorts of 2016. The objective was to have 600 teenage mothers that were eligible to the Program: 300 randomly selected to the control group (150 from the first cohort, 150 from the second) and 300 randomly selected to the treatment group (150 from the first cohort, 150 from the second).

The study comprises four measurements: baseline (before the implementation begins), a short term follow-up (6 months after baseline), a medium term follow-up (14 months) and long term follow-up-up (24 months after) follow-up studies are not guaranteed. The primary goal is to determine the treatment effect on both teenage mothers and their children, particularly on these dimensions: mothers subsequent pregnancy, sexual and reproductive health, educational attainment and labor market insertion; as well as in child nutrition, health and development.

The evaluation will provide policy guidance in regards to evidence-based intervention models that might assist vulnerable youth populations in their insertion into the labor market. Given the negative impacts of early motherhood and unplanned pregnancy on the general wellbeing of women and their offspring, the dynamics of teenage pregnancy is a central issue on the agenda of policy makers in the social, education and health sectors. Although general indicators of teenage pregnancy in Colombia show positive dynamics, there is huge variation across regions and a worrisome concentration of this phenomenon in particular municipalities of the Country. According to the World Bank’s World Development Indicators, between 2011 and 2014 the national teenage fertility rate (born alive per 1000 women between 15 and 19 years old) dropped from 58.9 to 51.7, while in some municipalities the opposite trend is observed. For instance, Cartagena (the city where the organization operates) is one of the cities with the highest incidence of teenage pregnancy: the city is in the fourth place in the ranking of all urban regions in Colombia. Moreover, the department of Bolivar (which circumscribes Cartagena) has an average fertility rate that is 10 percentage points above the national average.

External Link(s)

Registration Citation

Citation

Harker, Arturo and Bibiana Taboada. 2017. "Impact Evaluation of a Teenage Mother Program in Colombia." AEA RCT Registry. April 21. https://doi.org/10.1257/rct.2122-1.0

Former Citation

Harker, Arturo and Bibiana Taboada. 2017. "Impact Evaluation of a Teenage Mother Program in Colombia." AEA RCT Registry. April 21. https://www.socialscienceregistry.org/trials/2122/history/16809

Sponsors & Partners

There is information in this trial unavailable to the public. Use the button below to request access.

Request Information

Experimental Details

Interventions

Intervention(s)

The Juanfe Foundation is a non-profit organization that works to improve the quality of life of children and teens living in poverty in the city of Cartagena. The Foundation has two strategic objectives: reducing child mortality and empowering teenage mothers in extreme poverty, through comprehensive health, psychological and emotional care, and training in productive activities for their insertion into the labor market. The organization began operations in February 2002 and receives about 300 adolescent mothers per year.
The Foundation created a 360 degree model, which aims at transforming teenage mothers with low self-esteem into responsible mothers who own their lives, breaking the vicious circle of poverty. It is a 360 degree model because the intervention includes (i) sexual and reproductive health services, (ii) psychological care to improve the mothers’ self-esteem, build their confidence, develop their social skills, (iii) schooling (basic education), and (iv) vocational training. This comprehensive model is an innovative solution whose effect materializes in the medium and long run, when the mother begins to generate income and improve her quality of life. Rather than blaming the teen, the model works on variables that led to pregnancy in the first place: lack of education, poverty, emotional instability, and gender discrimination, among others risk factors. The model encompasses three phases.
Phase I (MA phase- lasts one semester) focuses on values and empowerment, for young mothers to develop their life plans and make informed choices about their future. It seeks to strengthen the self-esteem of each teenage mother with different activities throughout the semester. The education in sexual and reproductive health guides the mothers towards a responsible sexual life.
Phase II (ME phase- lasts two semesters) focuses on education and vocational training, enabling teens to return to secondary education or to begin a technical career to obtain a stable job. A percentage of them even make it to college. This phase continues to provide psychosocial support.
Phase III (OEE phase) contributes to the inclusion of adolescent mothers into the job market. Besides vocational training, mothers are trained on formulation and implementation of business plans. The Foundation identifies job opportunities in the city, and connects young mothers with formal labor market demand. Given the vagaries of job search, this phase has a variable duration.
The Foundation also provides child care for mothers who enter the programs. Annually it hosts 200 children, who are identified as those most in need of attention and monitoring. The objective of this initiative is to improve babies’ physical and psychomotor development, strengthen the mother-child bond and teach the young mothers how to care for their children. This also ensures greater retention of mothers in the program and avoids dropout.

Intervention Start Date

2016-01-04

Intervention End Date

2018-06-29

Primary Outcomes

Primary Outcomes (end points)

The evaluation will measure the impact of the intervention on a set of outcomes for the adolescent mothers and their children.
Outcome measures for the mothers will include: second pregnancy, use of family planning methods, health status, childcare practices, soft skills, educational attainment, employability, and income generation.
Outcome measures for children will include: height and weight, health status and socio-emotional development.

Primary Outcomes (explanation)

Tested scales will be used to asses outcomes related to childcare practices, soft skills and child development.

Secondary Outcomes

Secondary Outcomes (end points)

Secondary Outcomes (explanation)

Experimental Design

The experimental sample for the ongoing RCT has been constructed using an excess demand methodology. Given that the Teenage Mothers Programme selection process is based on an open call where applicants have to go through three different filters (cognitive tests, emotional tests and a face to face interview), to avoid salient selection biases in our impact estimates we had the necessity to construct the control group using the universe of individuals that were deemed as eligible. Evidently, if the counterfactual was constructed with the applicants that did not pass the filters, we would have important pre-existing differences between the individuals in the treatment and control groups. This is not the case in this RCT.

The experimental sample was constructed through the admissions process of two cohorts for 2016 using an excess demand method. The final objective was to have 600 teenage mothers that were eligible to the Teenage Mothers Programme: 300 randomly selected to the control group (150 from the first cohort, 150 from the second) and 300 randomly selected to the treatment group (150 from the first cohort, 150 from the second). The implementation of the RCT followed this implementation protocol:
1. Excess demand generation:
1.1. Promotion (publicity and information) campaign through local media and social assistance organizations.
1.2. The goal of this campaign is to have at least 900 teenage mother enlisted in the admissions process for the Programme.
1.3. After the admissions process we would have at least 300 teenage mothers that passed the above mentioned selection filters.
2. Baseline:
2.1. We designed a household survey to collect comprehensive information on: sociodemographic characteristics, health, education and labour outcomes, access to basic services, psychosocial characteristics, quality of the relationship between mother and child and sexual and reproductive health.
3. Lottery to define treatment and control groups:
3.1. Using the list of potential beneficiaries (i.e. Teenage mothers that are completely eligible), we use a random selection procedure to select them into the treatment group. This algorithm was written and implemented in the statistical software Stata.
3.2. Using the data collected in the baseline, we confirmed that there are no statistically significant differences across the treatment and control groups.
3.3. In addition, using this data we confirmed that the 2016 cohorts were similar to previous cohorts.

Experimental Design Details

Randomization Method

The randomization process was carried out in office using administrative and survey baseline data. The data was stratified by age groups, pregnancy status, and according to some specific quotas required by the Foundation's sponsors.

Randomization Unit

Randomization was carried out at the level of the individual adolescent mother.

Was the treatment clustered?

Experiment Characteristics

Sample size: planned number of clusters

600 adolescent mothers.

Sample size: planned number of observations

600 adolescent mothers.

Sample size (or number of clusters) by treatment arms

300 adolescent mothers in the treatment group and 300 adolescent mothers in the control group.

Minimum detectable effect size for main outcomes (accounting for sample design and clustering)

In the study's power calculation the probability of type 1 error (alpha) was 5% and, since units are not clustered, intra-class correlation is not an issue. Assuming that 20% of the variation of the outcome variables is explained by the available broad set of individual-level (e.g. age, school attendance, cognitive and socio-emotional skills) and household-level (e.g. family structure, income, dwelling characteristics) control variables at baseline, the minimum detectable effect size with a sample size of n=600 is 0.203 standard deviations (attaining a statistical power of 80%). Assuming that 40% of the variation of the outcome variables is explained by the set of control variables, the minimum detectable effect size with a sample size of n=600 is 0.177 standard deviations (attaining a statistical power of 80%). For this calculations we used the license-free software Optimal Design (http://hlmsoft.net/od/).

Supporting Documents and Materials

IRB