This paper examines factors associated with learning outcomes for children in Kakuma refugee camp, situated in northwestern Turkana County in Kenya. The authors document early literacy outcomes for children in lower primary schools in the refugee camp and explore the specific characteristics of refugee children and their settings (including country of origin, language of instruction, and the children’s expectations for their future) that are associated with learning outcomes. They also compare the learning outcomes of refugee children with those of Kenyan nationals outside the camps, with particular reference to host community children in Turkana County, which was ranked 45th out of 47 Kenyan counties in primary learning outcomes.
The analysis is based on a novel dataset consisting of the results of the Early Grade Reading Assessment (EGRA) for 732 children in lower primary classes (grades 1-3) representing all 21 schools in Kakuma refugee camp with lower primary classes, as well as two schools in the adjacent Kalobeyei refugee settlement. All schools follow the Kenyan curriculum, have both Kenyan and refugee teachers, and teach in English and Kiswahili. Nearly half (42 percent) of the assessed children were from South Sudan, 17 percent from Sudan, 9 percent from DRC, and 9 percent were from Somalia, with smaller proportions from other countries (e.g. Burundi and Eritrea).
- Learning outcomes for refugee children in Kakuma were exceedingly low, much worse than their counterparts in the rest of Kenya and even lower than those of disadvantaged host community children in Turkana County.
- Literacy outcomes differed among refugee children, depending on: (a) their country of origin; (b) language of instruction used at their school; (c) languages spoken at home; and (d) children’s expectations of a return to their country of origin. Somali refugees scored higher and South Sudanese refugees scored lower, possibly reflecting their length of stay in Kenya and varying exposure to education in Kenya and/or to English. The results are also suggestive, but not conclusive, that refugees’ expectations for the future shape early literacy learning. Refugee children who expected to remain in Kakuma for the following three years read less fluently in English, even though one might expect them to be more likely to invest in English language skills. Those who expected to return to their country of origin in ten years also read less fluently in English.
- The authors argue that their findings point to the urgent need to invest heavily in improving learning outcomes among refugee children, rather than focusing solely on their access to education. They suggest that two factors exacerbate the extremely poor learning outcomes for refugee children in Kakuma: (1) their differentiated learning needs, which may not be met entirely by the national education system; and (2) their marginalization. The findings suggest that a better understanding of students’ educational histories, their parents’ educational histories, and their sources of exposure to the languages of instruction, could usefully inform policy responses and appropriate instructional practices for students from different countries of origin. Additionally, the marginalization of refugees and host communities requires consideration of (and further research into) individual-level factors (poverty, family literacy) and school-level factors (teacher pedagogy, school environment) that contribute to learning outcomes.