Improving Adolescents Economic Well-Being through Economic and Psychological Empowerment in Low- and Middle-Income Countries

The study is a pilot clustered randomized control trial with 4 arms: self-regulation, economic, a combined arm incorporating both self-regulation and economic interventions, and a control group

2024-05-23

2024-11-07

We collect data in the baseline and subsequent follow ups for four primary outcomes: education beliefs, financial skills, negotiation skills, and consumption

We construct three measures for education beliefs and aspirations:
• Beliefs-Labor: This variable was constructed based on the following question: “Which of the following do you think is true for adolescents who complete secondary schooling?” The responses are: 1-They are more likely to end up working in a paid job than those who do not complete it; 2-They are equally likely to end up working in a paid job than those who do not complete it; and 3-They are less likely to end up working in a paid job than those who do not complete it. An indicator variable is constructed to equal one if the respondent selects option 1 and zero otherwise.
• Beliefs-Earnings: This variable was constructed based on the following question: “Which of the following do you think is true for adolescents who complete secondary schooling?” The responses are: 1-They are likely to earn more in their lifetime than those who do not complete it; 2-They are likely to earn similarly in their lifetime than those who do not complete it; and 3-They are likely to earn less in their lifetime than those who do not complete it. An indicator variable is constructed to equal one if the respondent selects option 1 and zero otherwise.
• High Edu Aspiration: This variable was constructed based on the following question: “How much do you expect to complete?”: The responses are: 1-Elementary school (grade 5) [Col]; 2-Secondary school (grade 9) [Col]; 3-High school (grade 11) [Col]; 4-Technician or technologist (e.g. studying at SENA) [Col]; 5-University degree [Col]; 6-Specialization or master's degree [Col]; 7-SEE or below [Ne]; 8-10+2 [Ne]; 9-Bachelors [Ne]; (x) 10-Masters or above [Ne]; 11-Primary (till grade 7 or equivalent) [SA]; 12-Secondary (e.g. high school/technical or vocational school) [SA]; and 13-University/tertiary (e.g. University/Technikon, College) [SA]. An indicator variable equal to one if the respondent selects option 5 or higher in Colombia, 8 or higher in Nepal, and 12 or higher in South Africa and zero otherwise.

We construct three measures for financial skills:
• Budgeting: An index constructed by standardizing three indicators related to: (i) having a budget or spending plan; (ii) no difficulty tracking spending; and (iii) no difficulty avoiding wasteful spending
• Spending control: An index constructed by standardizing three indicators related to (i) being able to avoid spending money; (ii) having saved money each month for past months; and (iii) having control over whether or no can spend/save money
• Saving control: An index constructed by standardizing three indicators related to (i) no difficulty saving money over the past two months; (ii) being able to buy things needed after saving; and (iii) no need to borrow after saving

We construct two measures for negotiation skills, both winsorized at the 99th percentile:
• Schoolwork: Average hours spent per day by participants on schoolwork and studying at home in the last week
• Chores: Average hours spent per day by participants on household chores in the last week
• Schoolwork vs Chores: A ratio of hours spent per day on schoolwork versus hours spent per day on chores

We construct measures for consumption:
• Personal consumption: An index constructed by standardizing measures of consumption related to clothing, shoes, beauty, and jewelry
• School-related items: An index constructed by standardizing measures of consumption related to books and school supplies

We collect data in the baseline and subsequent follow ups for two downstream outcomes: school adverse events and school attendance.

We construct three measures for school adverse events in the past 6 months:
• Failed: An indicator variable equal to one if participant has failed a full academic year and zero otherwise
• Dropped out: An indicator variable equal to one if participant has dropped out of school for any reason and zero otherwise
• Kicked out: An indicator equal to one if participant have been kicked out of school and zero otherwise
• Adverse events: An index constructed by standardizing three previous measures

We construct three measures for school attendance in the past 4 weeks:
• Skip classes: An indicator equal to one if participant has skipped some classes and zero otherwise
• Skip day: An indicator equal to one if participant has skipped a whole day of school and zero otherwise
• Late school: An indicator equal to one if participant has arrived late for school and zero otherwise
• School attendance: An index constructed by standardizing three previous measures and zero otherwise

4-arm cluster Randomized Controlled Trial

We randomly allocated schools (clusters) to four arms in each site using a 1:1:1:1 ratio

School

Yes

24 schools

628 adolescents

6 schools self-regulation, 6 schools economic intervention, 6 schools combined intervention, and 6 school control group

Analysis Plan_Improving Adolescents Economic Well-Being in LMICs.pdf