How are children learning? Uwezo assessment findings from our partners in Tanzania


In Tanzania, Firelight is working to encourage, strengthen, and replicate or scale innovations in early learning among community-based organizations to improve children’s school readiness, literacy, and numeracy outcomes. Each of Firelight’s grassroots partners under this program takes a holistic and innovative approach to increasing community involvement in children’s early learning. Supported by The William and Flora Hewlett Foundation, the overarching goal of our Grassroots Innovations in Early Learning Initiative is to improve reading, writing, and arithmetic levels for children in the early years of education. In mid-2015, Firelight supported the collection of learning outcome data from a sample of children supported by our grantee-partners’ programs. We used the Uwezo assessment tool, administered through mobile data collection technology.

Read more about our experience using mobile technology here. And about the Uwezo tool here.

We completed assessments with 266 children in Tanzania, from the Dar es Salaam, Kilimanjaro, and Mwanza regions. Children ranged in age from 5 – 14 years. Of the children for whom grade level information was available, most were in grade 3 (63%), with the remainder in grades 1 and 2.

The Uwezo assessment is intended to assess learning outcomes for students in grade 3, by testing them in 3 areas: Kiswahili, Numeracy, and English. Passing is considered to be the ability to meet grade 2 standards for each subject – that is, reading a story in Kiswahili and English and doing multiplication.

Key Findings

  1. Compared to a national sample, the Firelight-sampled grade 3 students performed similarly in Kiswahili and Numeracy, but had lower pass rates in English

[visualizer id="6947"]


We compared Firelight’s results with the latest published results from the 2013 national Uwezo assessment in Tanzania, as a benchmark. Our comparisons indicated that:

  • 52% of the Firelight-sampled Grade 3 students passed the Kiswahili literacy test in 2015, comparable to or slightly higher than the 45% who did from the national 2013 sample;
  • 7% of the Firelight-sampled Grade 3 students passed the English literacy test in 2015, lower than the 19% who did from the national 2013 sample; and
  • 29% of the Firelight-sampled Grade 3 students passed the numeracy test in 2015, comparable to the 31% who did from the national 2013 sample.

While our sample of children performed similarly to the national sample in Kiswahili and in Numeracy, they had lower pass rates in English than the national sample. This could be due to the fact that our grantee-partners tend to work with the poorest performing students, providing them with targeted support. Thus, Firelight’s sample is not representative of the wider student population.

Nonetheless, the Uwezo findings provide us and our partners with important information about how students are learning, and about areas where improvement is needed. However, regardless of performance level, less than half of students included in both Firelight’s and the national sample are performing at grade level. These findings put into question the quality of learning environments and preconditions in place for children to thrive at school.

  1. Children experience limited educational interactions at home

The student assessment was accompanied by a household survey completed by the child’s primary caregiver. Caregivers were asked about interactions with their children in support of the child’s learning and education.

[table id=1 /]

Half of the households did not have any books, and the other half generally had few books (94% of households had fewer than 10 books). In addition, the majority of caregivers (74%) reported that they had not read with their child in the past week, while a quarter (26%) reported that they had.

However, the majority of caregivers (68%) reported that they had checked their child’s homework in the past week.

And, over a third (39%) reported having recently had a discussion with the child’s teacher, and over two-thirds (68%) reported attending a parents’ meeting at the school.

These findings identify areas of strength as well as areas needing additional work for our grantee-partners in their work supporting children’s early learning. Parents and caregivers are critical influencers to their child’s learning, especially at home, and efforts need to be made to create more inclusive environments for parents to actively contribute to learning.

In the next phase of our initiative, we’re using these findings to develop targeted approaches to address gaps in children’s learning both at home and in school.