SEED 2.0 - RCT at Scale

How to cite

Chioda, Laura; Gertler, Paul. 2025. "SEED 2.0 - RCT at Scale." AEA RCT Registry. AEARCTR-0015810. https://doi.org/10.1257/rct.15810-1.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Laura Chioda* , Paul Gertler

* Primary investigator

Registry ID
AEARCTR-0015810
DOI
https://doi.org/10.1257/rct.15810-1.0
First registered
Last updated
Status
In development
Trial dates
2025-03-08 – 2028-06-15
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/15810

Abstract

The at-scale Randomized Controlled Trial (RCT) involving 8,500 Ugandan youth will generate rigorous evidence on impacts at scale of a tested and innovative youth skills and entrepreneurship training, Skills for Effective Entrepreneurship Development (SEED). The at-scale RCT aims at gaining insights on the continued effectiveness of SEED at-scale and the added value of including a digital business skills module in combination with a Generative AI tool. The SEED-tailored AI-powered virtual tutor will be available during training and as youth embark on their entrepreneurial journeys.
SEED is a 3-week in-residence mini-MBA program that was first implemented as an RCT in 2013 (by UC Berkeley), targeting youth at the end of secondary school (S6), which is earlier in life than traditional training and a critical stage for the skills development. Four years post-intervention, youth in the treatment group were more likely to start enterprises and more successful in ensuring business survival. Ten years post-intervention, a very conservative measure of the present discounted value of SEED-induced extra business earnings. The SEED intervention and its scale-up speak to a pressing policy priority. Entrepreneurship is vital for job creation and economic growth, especially in Africa, with governments investing heavily in programs designed to expand entrepreneurial skills and spending over US $1 billion annually on entrepreneurship training (McKenzie 2021). T

Intervention

The an at-scale Randomized Controlled Trial (RCT) involving ~8,500 Ugandan youth will generate rigorous evidence on impacts at scale of a tested and innovative youth skills and entrepreneurship training, Skills for Effective Entrepreneurship Development (SEED). The at-scale RCT aims at gaining insights on the continued effectiveness of SEED at-scale and the added value of including a digital business skills module in combination with a Generative AI tool. The SEED-tailored AI-powered virtual tutor will be available during training and as youth embark on their entrepreneurial journeys.
SEED is a 3-week in-residence mini-MBA program that was first implemented as an RCT in 2013, targeting youth at the end of secondary school (S6), which is earlier in life than traditional training and a critical stage for the skills development. Key elements of SEED that yielded large and sustained effects 10 years post-intervention are preserved and offer confidence for its future success. SEED was designed/implemented as a high-quality intervention. The curriculum was informed by rigorous economics and psychology evidence; it targeted youth at the end of secondary school (S6), which is earlier in life than traditional trainings. SEED was designed to promote active, participatory learning.

Experimental design

Study Design. 8,500 youth from a nationally representative, gender-stratified sample will be randomly assigned as follows:
● Treatment 1 (T1): 2,853 S6 youth, residential SEED (~3 weeks)
● Treatment 2 (T2): 2,853 S6 youth, residential SEED + Digital and AI tool (~3 weeks)
● Control Group (C): 2,853 S6 youth

Additional stratification dimensions will include (i) geographical areas, capturing youth that can be served by the program to reach adequate, cost-effective operational-scale for implementation; and (ii) distance from the expected venues to which youth will be assigned. Namely, youth will be assigned to two distance-bands: 0-30km and 30-60km.

Design details

Randomization method
Computer
Randomization unit
Individual.
Clusters
NA
Observations
8,500 youth
Arms
Treatment 1 (T1): 2,853 S6 youth, residential SEED (~3 weeks) ● Treatment 2 (T2): 2,853 S6 youth, residential SEED + Digital and AI tool (~3 weeks) ● Control Group (C): 2,853 S6 youth
Primary outcomes
- Hard and soft skills upgrading (knowledge tests; task-based measures; socio-emotional and personality traits) - Education investments (e.g., post-secondary education, field of study). Note in the original SEED, no trade-offs were observed between entrepreneurial effort/success and educational investments. - Business starts, active businesses, and entrepreneurial experience - Business characteristics (formality, employment, number of employees, partnerships, access to finance, etc.) - Managerial practices and tech adoption, including businesses’ online presence - Businesses’ financial performance (revenues, capital) - Earnings (business earnings, wage earnings, total earnings)
Secondary outcomes
- Soft-Skills and Personality Traits: As in the past, we plan on including a battery of soft-skills and personality trait scales(Big Five, grit, self-efficacy, prosociality, and mental health measures (GAD/PHQ), but we hope to validate soft-skills via task-based measures, as well as lab-in-the-field experiments). - Comprehensive measures of well-being (anxiety, depression, expectations) - Social spillovers: fertility, intimate partner violence, gender norms, gender-empowerment.

Cited by

No citation data available for this registration yet.

Keywords