Stories that Shape Us: Evidence from a Gender-Empowering Literacy Program

The intervention integrates foundational reading instruction with gender-empowering messaging in a school-based literacy program in Botswana. This is delivered during the remedial hour in government primary schools, the program was implemented in partnership with Youth Impact under an MOU with the Ministry of Education. Instruction was targeted to students’ literacy levels using the Teaching at the Right Level approach.

Group A – Status Quo: Teachers developed their own reading lessons using optional guidelines and a resource kit. Materials primarily included descriptive stories on everyday topics (e.g., family, school, animals) without explicit empowerment content.

Group B – Empowering Stories: Teachers were encouraged to use a resource kit with stories embedding messages of gender equality, leadership, perseverance, and non-stereotypical roles. Narratives emphasized agency and role models (e.g., female leaders). Exposure was concentrated among students working at paragraph and story levels, with minimal exposure for letter- and word-level groups.

2024-01-15

2025-07-30

Economic agency and aspirations – binary variable on students’ agreement with the statement “Girls can do any job they want”

Specific job aspirations - binary from response to open-ended question: “What job do you want to do when you grow up?” Responses will be cleaned and categorized into the occupations with greater than 1% responses.

Economic agency and aspirations – binary variable on students’ agreement with the statement “Boys can do any job they want”

Belief on student ability – binary variable on students’ agreement with the statements “Girls are smarter", "STEM is for girls" and "Math is for girls"

Learning outcomes - continuous variable of students literacy level using ASER assessment.

We conducted a randomized experimental evaluation of empowering stories integrated into a primary school literacy curriculum across six of Botswana’s ten regions. Over five rounds of A/B testing between January 2024 and August 2025, 100 public primary schools (5,616 students in grades 3–5) were randomly assigned to either the status quo content (Group A) or empowering stories (Group B). Schools and implementing staff were blinded to their treatment assignment until training, held one to two days before implementation.

Group A materials featured descriptive stories about households or hobbies, while Group B included empowering, gender-progressive narratives. The intervention was delivered through daily one-hour literacy sessions within a teaching at the right level framework each school term. Only students in paragraph groups and story groups based on baseline learning levels were exposed to empowering stories.

Our analysis draws on anonymized administrative and survey data collected at baseline and endline for each round, including student demographics (e.g., gender, grade) and measures of empowerment outcomes.

Not available

STATA generated randomization

School

Yes

100 schools

5616 students

A: 2,728 B: 2,888 students