GirlsRead! Zambia DREAMS Innovation

The primary aim of this study is to determine through a randomized evaluation design whether an e-reader intervention embedded in a safe space platform with a community engagement component will bolster literacy skills among adolescent girls, enhance critical thinking skills, improve results on the Grade 7 Composite Exam, and increase progressing to secondary school. A host of contextual factors increase adolescent girls' risk for premature school leaving and subsequent HIV infection in Zambia and evidence suggest that female education, particularly secondary school, may act as a "social vaccine" to prevent the spread of HIV.

2017-03-20

2017-10-31

Primary Outcome:
Literacy skills: measured at baseline and endline (approximately 12 months later) using an assessment developed in English for 6-16 year olds by the People's Action for Learning (PAL) Network for Anglophone African countries. The literacy assessment is a simple diagnostic of individual progress in reading that tests letter recognition, reading simple words and comprehension. However, because the PAL assessments are pitched at a low level they may generate insufficient variability. Thus we have added some literacy and numeracy questions from a prior Zambian Grade 7 Composite examination as well as some questions from the Zambian Grade 3 and 5 English textbooks.

Secondary Outcomes:
1. Time spent reading: measured at baseline and endline using a structured questionnaire by asking participants how much time they usually spend reading, how much time they spent in the last day and how much time in the last 7 days.
2. Score on the Grade 7 Composite exam: With authorization from the Ministry of General Education we will collect Grade 7 Composite Exam scores for all participants interviewed at baseline who are sitting for the exam in October 2017.
3. Progression to secondary school: measured at endline using a structured questionnaire by asking participants whether they are enrolled in secondary school.
4. Numeracy skills: measured at baseline and endline using an assessment developed in math for 6-16 year olds by the People's Action for Learning (PAL) Network for Anglophone African countries. The math assessment measures students' foundational skills including number identification, counting, word problem solving, addition, subtraction, multiplication and division. As with literacy, we have added some numeracy questions from a prior Zambian Grade 7 Composite examination as well as some questions from the Zambian Grade 3 and 5 mathematics textbooks.
5. Attitudes and knowledge:
a. gender attitudes: measured through administering GEM scale questions to participants as well as additional questions about women's rights at baseline and endline
b. reproductive health knowledge: measured through a series of questions to participants on HIV and pregnancy risk as well as knowledge about where services can be obtained at baseline and endline
c. self-efficacy: measured through a generalized self-efficacy scale administered to participants as well as additional questions on expressing opinion and confidence asking questions at baseline and endline
6. Sexual behavior: measured at baseline and endline using ACASI by asking participants about sexual initiation, condom use, and partners.

Exploratory Outcomes:
1. critical consciousness: measured through five questions to participants on women's experiences in Zambia regarding gender equity in employment, abuse from male partners, early marriage, and speaking out about injustice, at baseline and endline
2. perpetration of school violence: measured through five questions about participant name calling, teasing, stealing, physical violence etc… directed at other students at baseline and endline

GirsRead! is a multi-center interventional study designed to enhance learning and increase progression to secondary school among adolescent girls in grade 7, the last year of primary school, in three districts located in two provinces in Zambia; the evaluation will use a 3-arm randomized cluster trial design, with structured surveys administered before and after the intervention to adolescent girls in Grade 7, the last year of primary school.
The three arms are:
Arm A: Community engagement activities, safe space groups and e-readers.
Arm B: Community engagement activities, safe space groups
Arm C: No activites

The study will be structured as follows:
Arm A: Community engagement activities, safe space groups and e-readers. Arm A sessions will devote approximately 30-45 minutes to the safe space curriculum and 30-45 minutes to discussing content on the e-reader.
Arm B: Community engagement activities, safe space groups. Arm B sessions will devote approximately 30-45 minutes to the safe space curriculum and 30-45 minutes to group review of the school materials in order to determine if time devoted to reading material considered to be entertaining improves literacy skills and performance on the Grade 7 Composite Exam more than time devoted to academic review.
Arm C: No activities.

Each safe space session is expected to last approximately 60-90 minutes. Holding the "dosage" of the safe space groups constant will allow us to reject the alternative explanation that smiply more time in the safe space groups could explain improved outcomes.
Arm C: No activites

Selected schools will be randomly assigned to arms by generating a random number for each school in Microsoft Excel, then sorting by that random number and assigning arms A, B, or C in sequence.
Rosters from the selected schools were imported into STATA and a random number was assigned to each girl. Within each school, we sorted by the random number and attempted to interview the first 38 eligible girls.

The trial is randomized at the school level. Within a school, a random number was assigned to each girl; as noted above, we sorted by the random number and will attempt to interview the first 38 eligible girls.

Yes

The study will have a sample size of 36 schools.

The total number of participants is approximately 1,152 females enrolled in Grade 7

There will be 3 arms, with 12 schools per arm and approximately 32 participants per school,

With a sample size of 1152 (36 schools - 12 per arm - and approximately 32 participants per school), power of 0.80, an alpha of 0.05, and intra cluster correlation of 0.01 we will be able to detect an increase of 11 percentage points from 27% to 38% for literacy (based on the percentage of pupils in grade 6 that could read for meaning in the SACMEQ Zambia assessment), and 9 percentage points from 78% to 87% for progression to secondary school (based on percentage of grade 7 students in 2007 Zambia DHS who progressed to secondary school).