Afya Rights Initiative
Afya Rights
Initiative
All documents
Research study

Girls' Retention Study, Nyando 2025

A mixed-methods study of girls' retention in upper primary and secondary schools in Nyando sub-county, with cohort cascade, ward-level patterns, and what works.

Published August 2025Afya Rights InitiativeKisumu, KenyaARI-RS-2025-04v1.0≈ 36 min readCC BY-NC 4.0
Foreword

The girls who shared their stories with this study were generous with details that, for many, were painful to revisit. They told us about chores at 04:30, about walks to school past sugar-cane fields, about teachers who saw them and teachers who did not, about boys who waited at the bus stop, about the term that began and the term that did not, about the cousin who married at fifteen, about the auntie who took her in.

We have done our best to honour those stories with care. Identifying details are changed in every case study. The quantitative findings are reported with the limitations they deserve. The recommendations are addressed to the people closest to the problem, head teachers, ward administrators, sub-county education officers, the Department of Education and Skills Development, ARI itself, and to the national policy actors who set the frame.

A note of honesty: this study began as a small fieldwork exercise to inform ARI's own programme design. Two months in, we recognised that the patterns emerging deserved a wider audience. The methodology was adjusted accordingly. The result is not a peer-reviewed paper. It is a working study, transparent about what it can and cannot say.

- The Afya Rights Initiative team

Section 1

Executive summary

1,432
Girls tracked across nine Nyando schools, 2022–2025
18.4%
Cohort drop-out rate over the three-year window (ARI estimate)
9.1–28.7%
Ward-level range of cohort drop-out rates
73
Key informant interviews and 28 focus group discussions
Section 2

Why Nyando, why now

Nyando sub-county comprises five wards: Awasi/Onjiko, Ahero, Kabonyo/Kanyagwal, East Kano/Wawidhi, and Kobura. The sub-county's economy is dominated by smallholder agriculture and sugar-cane farming, with seasonal labour patterns that interact tightly with school attendance. Nyanza region has consistently recorded one of the highest rates of adolescent childbearing in Kenya: KDHS 2022 reports approximately 21% of girls aged 15–19 in Nyanza have begun childbearing, the highest of any region[1].

The Ministry of Education's 2023 Basic Education Statistical Booklet[2] documents continued progress on girls' enrolment but persistent gaps in retention and transition, particularly at the primary–junior secondary boundary that falls in the centre of the period covered by this study.

Section 3

Research questions

  1. What is the cohort retention rate of girls in upper primary and secondary schools in Nyando sub-county over a three-year window (2022–2025)?
  2. How does retention vary across the five wards of Nyando, and across the nine schools in the sample?
  3. What are the most cited drivers of drop-out, in the words of girls who left school and the adults around them?
  4. What practices and supports are visible in schools that retain girls at materially higher rates than peer schools?
  5. What are the implications for county budgets, school-level practice, and ARI's own programme design?
Section 4

Methodology

Section 5

The sample

Table 1. Sample frame at baseline, by ward and school level
WardSchools sampledUpper primary girlsJunior secondary girlsSenior secondary girlsTotal
Awasi/Onjiko2142860228
Ahero21389278308
Kabonyo/Kanyagwal2156840240
East Kano/Wawidhi1886471223
Kobura2174142117433
Total96984682661432
Source: ARI cohort registers, January 2022 baseline.
Section 6

The cohort cascade

Figure 1. Cohort cascade, what happened to 1,432 girls in three years
1. Girls enrolled at baseline (Jan 2022)1,432
2. Girls still enrolled at end of year 11,391(97% of start)
3. Girls still enrolled at end of year 21,297(91% of start)
First major drop, concentrated at the upper primary → junior secondary transition.
4. Girls still enrolled at end of year 3 (May 2025)1,168(82% of start)
Endline cohort.
5. Girls who returned after temporary absence36(3% of start)
Documented re-entry, almost all via head-teacher initiative.
6. Girls confirmed out of school at endline264(18% of start)
Source: ARI cohort tracking, 2022–2025.

Of the 264 girls confirmed out of school at endline, 81 were documented as having entered marriage; 96 as having had a child; 47 cited financial pressure or household labour; and the remaining 40 cited a combination of factors. Categories overlap and do not sum.

Section 7

Drivers of drop-out (in the words of girls)

Quantitative tracking tells us that girls leave. Qualitative work tells us why. The chart below summarises the most cited drivers across the 14 in-depth life-history interviews and the 28 focus group discussions, weighted by frequency.

Figure 2. Most cited drivers of drop-out, across qualitative work (multiple-mention permitted)
Menstrual hygiene management gaps (cost, access, dignity)≈64%
Distance, safety, and harassment on school journey≈58%
Household care responsibilities (sibling care, farm labour)≈49%
Adolescent pregnancy and early union≈44%
School fees / levies / uniform costs≈41%
Non-supportive school environment after pregnancy≈32%
Mental health and undiagnosed learning support needs≈18%
Source: ARI qualitative data, 2024–2025. Indicative.

You can manage one term without proper pads. Maybe two. The third term, you start to make a calculation: do I miss class this week, or do I stop coming altogether? It is not a free choice. The school is not asking the question.

Form 2 girl, focus group discussion, Ahero ward, March 2025
Section 8

Ward × school retention table

Table 2. Ward × school cohort retention, baseline to endline
WardSchool (de-identified)Baseline nEndline nRetained %Notes
Awasi/OnjikoSchool A1167968.1%Highest single-school drop-out in sample
Awasi/OnjikoSchool B1128475.0%
AheroSchool C16814485.7%Tuseme-style club active
AheroSchool D14011682.9%
Kabonyo/KanyagwalSchool E1249879.0%
Kabonyo/KanyagwalSchool F1169279.3%
East Kano/WawidhiSchool G22319788.3%Strong head-teacher re-entry practice
KoburaSchool H21918484.0%Tuseme-style club active
KoburaSchool I21417481.3%
Source: ARI cohort tracking, 2022–2025. Schools de-identified at school-name level by request.
Section 9

Four girls, four pathways

Case study

Ruth, 14

Awasi/Onjiko ward
Drop-out: 2024 Term 2
Situation
Ruth was a top-quartile student in upper primary. Her father died at the end of 2023; her mother began a daily 90-minute walk to a sugar-cane processing site. Ruth assumed sibling care. By Term 2 of 2024 she was attending two days per week.
ARI response
The class teacher noticed and referred Ruth to ARI's community partner. A small monthly stipend (KES 1,800) supported a neighbour to take on sibling care during school hours. The school's Tuseme club paired Ruth with a peer for catch-up.
Outcome
Ruth re-entered full-time attendance for Term 3 of 2024 and completed the year. She is currently enrolled in Form 1. The stipend has been transitioned to a community save-and-loan group.

Composite case study, anonymised. Names, ages and identifying details have been changed to protect the individuals concerned.

Case study

Mercy, 16

Awasi/Onjiko ward
Drop-out: 2023 Term 3
Situation
Mercy became pregnant at the end of 2023. She told her aunt and was supported through the pregnancy, but did not return to school in 2024 due to a combination of lack of childcare, embarrassment, and a school environment she described as "judgemental".
ARI response
ARI partnered with the head teacher of a nearby school to facilitate a re-entry plan in line with the Ministry of Education's 2020 re-entry guidelines[4]. A community grandmother joined a small daycare cluster.
Outcome
Mercy returned in Term 1 of 2025. Attendance is sustained through Term 2. She has been paired with a re-entry navigator from the Tuseme club.

Composite case study, anonymised. Names, ages and identifying details have been changed to protect the individuals concerned.

Case study

Faith, 13

Kobura ward
Drop-out averted: 2024 Term 1
Situation
Faith began missing school during her menstrual period in 2023. By early 2024 the absences were stretching to four days per cycle. Her grandmother was unaware; Faith was embarrassed.
ARI response
The school's life-skills teacher (in line with KICD guidance[5]) ran a Tuseme session that made the issue discussable. A modest pad supply was added to the school's existing UNICEF-aligned MHM programme[6].
Outcome
Faith's absences fell to under one day per cycle by mid-2024 and to zero by 2025 Term 1. She is in the top three of her class.

Composite case study, anonymised. Names, ages and identifying details have been changed to protect the individuals concerned.

Case study

Achieng, 17

East Kano/Wawidhi ward
At endline: Form 4
Situation
Achieng walked 6.4 km each way to her secondary school. The walk passed two sugar-cane fields where harassment was regular and one stretch with no other walkers in early morning hours.
ARI response
ARI's Community Safe Journeys pilot[7] organised a rotating walking group of six girls, with a community boda-boda contact for emergencies. The school re-set its first-period start by ten minutes to align with the group's arrival.
Outcome
Achieng completed Form 3 with no recorded absence due to safety. She is on track to sit her KCSE this year.

Composite case study, anonymised. Names, ages and identifying details have been changed to protect the individuals concerned.

Section 10

Nyando vs national benchmarks

Figure 3. Selected girls' education indicators, Nyando vs national benchmarks
Nyando cohort (this study)National (KDHS 2022 / MoE 2023)
Adolescent (15–19) childbearing rate (region)
21 %
15 %
Cohort drop-out across 3-year window (girls)
18 %
11 %
Schools with active girls' club
22 %
18 %
Re-entry following pregnancy (within 18 months)
38 %
31 %
Source: ARI cohort, KDHS 2022, MoE Basic Education Statistical Booklet 2023. Comparators are nationally aggregated and provided for orientation.

Nyando is consistently above national benchmarks on the adverse indicators and roughly at parity on the supportive indicators. The implication is straightforward: the supports that exist are working at roughly national average; the headwinds are stronger. The policy lever is to strengthen the supports against the local headwind, not to import a generic national average response.

Section 11

What works (four protective signals)

Across the nine schools and the wider community work, four protective signals appeared repeatedly in the schools and contexts that retained girls at higher rates. They are not new. They are consistent with the published evidence base on what works for girls' education[8].

  1. A girls' club that meets weekly. Tuseme-style clubs[9] with a designated patron teacher, a running peer-mentor model, and at least one annual life-skills module aligned to KICD guidance.
  2. A reliable MHM supply. A buffer stock of pads at the school, integrated into the school's existing health stock, with a low-friction request pathway that does not require a girl to ask the male teacher for them.
  3. A named, supported re-entry navigator. One teacher (often the deputy head) who knows the 2020 re-entry guidelines[4] and is empowered to operate them, including making the home visit when needed.
  4. A safe-journey practice. Walking groups, a shifted first-period bell, a community contact list, small, specific arrangements that displace the largest share of harassment-related absence.
Section 12

Recommendations

Table 3. Recommendations for the FY 2025/26 cycle and beyond
#RecommendationOwner
1Resource a Tuseme-style girls' club at every public school in Nyando, with a costed patron-teacher honorariumSub-County Director of Education, DESD
2Standardise an MHM buffer stock at every school, with a low-friction request pathwayDESD, school heads
3Name and train a re-entry navigator in every secondary school in NyandoDESD, MoE liaison
4Pilot a Community Safe Journeys network across the two highest drop-out wardsARI + community administrators
5Co-design with the Department of Health a school-level MHM commodity line in the county budgetDoH, DESD
6Adopt cohort retention as a sub-county-level reported indicatorDESD, Health Committee
7Revisit the school transport allowance question in light of Section 11 finding (4)County Treasury
Source: ARI.
Section 13

Limitations and what we are not saying

  • Nine schools is not a representative sample of Nyando, and Nyando is not Kisumu. Generalisation should be cautious.
  • School-level differences in retention are correlated with the presence of girls' clubs, but causal claims would require a larger comparison group and ideally a quasi-experimental design[10][11].
  • Re-entry data is documented through head-teacher records, which under-count silent re-entry.
  • Qualitative data is weighted toward the voices of girls who could be reached. Girls who had left the area are under-represented.
  • All financial and economic figures cited as ranges are indicative and labelled as such.
Annex A

Survey instrument (summary)

Baseline and endline surveys covered:

  • Demographic and household composition.
  • School attendance, repetition, and academic confidence.
  • Menstrual hygiene management, access, cost, school adequacy.
  • Distance, journey time, and reported harassment on the school journey.
  • Household care responsibilities (a 7-day recall instrument).
  • Sources of support: school, family, community, ARI/partner.
  • Aspirations, life skills, and self-reported knowledge of SRHR services available locally.
Annex B

Sample frame and school-level details

The sample frame was constructed from 2021 class registers, stratified by ward and school level. Schools were selected purposively to ensure ward coverage and the inclusion of both schools with and without an active girls' club at baseline. The full school-level table is available on request. For privacy reasons we do not publish school-level identifiers in this report.

Reference

Glossary

Cohort retention
The proportion of an originally identified group of pupils who are still enrolled at a later observation point.
DESD
Department of Education and Skills Development (Kisumu County).
KCSE
Kenya Certificate of Secondary Education.
KDHS
Kenya Demographic and Health Survey.
KICD
Kenya Institute of Curriculum Development.
MHM
Menstrual Hygiene Management.
MoE
Ministry of Education.
Re-entry
Return to formal education following a documented absence (e.g. after pregnancy).
Tuseme
Swahili for 'let us speak'. A girls'-club approach pioneered by FAWE focusing on voice, agency, and life skills.
Ward
Sub-unit of a Kenyan sub-county; the smallest electoral administrative unit.
Acknowledgements

ARI thanks the head teachers, deputy head teachers, life-skills teachers, patron teachers, parents, guardians, ward administrators, and Sub-County Director of Education staff who supported the fieldwork. We thank the community organisations and faith leaders in Nyando who created safe space for the discussions. We thank the girls and young women themselves, who gave their time, their stories, and their judgement. The responsibility for the analysis and the recommendations is ours alone.

References

Bibliography

  1. [1]Kenya National Bureau of Statistics et al. (2023). Kenya Demographic and Health Survey 2022. KNBS and ICF.
  2. [2]Ministry of Education, Kenya (2023). Basic Education Statistical Booklet 2023. State Department for Basic Education, Nairobi.
  3. [3]Bandiera, O., Buehren, N., Burgess, R., Goldstein, M., Gulesci, S., Rasul, I., & Sulaiman, M. (2020). Women's Empowerment in Action: Evidence from a Randomised Controlled Trial in Africa. American Economic Journal: Applied Economics, 12(1), 210–259.
  4. [4]Ministry of Education, Kenya (2020). National Guidelines for School Re-entry in Early Learning and Basic Education. Nairobi.
  5. [5]Kenya Institute of Curriculum Development (2021). Life Skills and Values Education Framework. Nairobi: KICD.
  6. [6]UNICEF (2022). Re-imagining Girls' Education: Solutions to Keep Girls Learning in Emergencies. New York: UNICEF.
  7. [7]Afya Rights Initiative (2024). Community Safe Journeys Pilot, Internal Evaluation. Kisumu (mimeo).
  8. [8]Brookings Institution (2022). What Works in Girls' Education: Evidence for the World's Best Investment. Washington, DC.
  9. [9]Forum for African Women Educationalists (2020). Tuseme Empowerment Programme Manual. Nairobi: FAWE.
  10. [10]Evans, D. K., & Yuan, F. (2022). What we learn about girls' education from interventions that do not focus on girls. World Bank Economic Review, 36(1), 244–267.
  11. [11]Duflo, E., Dupas, P., & Kremer, M. (2015). Education, HIV and Early Fertility: Experimental Evidence from Kenya. American Economic Review, 105(9), 2757–2797.
  12. [12]UNESCO (2022). Leave No Child Behind: Global Report on Boys' Disengagement from Education. Paris: UNESCO.
  13. [13]Ministry of Health, Kenya (2015). National Adolescent Sexual and Reproductive Health Policy. Nairobi.
  14. [14]J-PAL Africa (2021). Reducing teenage pregnancy: Evidence from a programme in Kenya. Abdul Latif Jameel Poverty Action Lab.
  15. [15]Wangia, J., Mwingi, M., & Otieno, S. (2021). School re-entry after adolescent pregnancy in western Kenya: Lessons from Nyanza counties. East African Medical Journal, 98(4), 190–198.
  16. [16]County Government of Kisumu (2023). Kisumu County Integrated Development Plan III, 2023–2027.
  17. [17]Kenya National Bureau of Statistics (2023). Kenya Population and Housing Census Volume IV. Nairobi.
  18. [18]Afya Rights Initiative (2025). County SRHR Financing Brief 2025 (ARI-PB-2025-03). Kisumu.
  19. [19]Afya Rights Initiative (2025). GBV Response Capacity in Kisumu Level 4 Facilities (ARI-CA-2025-02). Kisumu.
  20. [20]WHO, UNESCO, UNFPA, UNICEF, UN Women, UNAIDS (2018). International Technical Guidance on Sexuality Education, revised. Paris: UNESCO.
  21. [21]Cooper, P., Tomlinson, M., et al. (2020). Cash transfers and adolescent education in sub-Saharan Africa: a systematic review. Journal of Adolescent Health, 67(3), 312–322.
About Afya Rights Initiative

Afya Rights Initiative (ARI) is a youth-led community organisation based in Kisumu, Kenya, working at the intersection of sexual and reproductive health rights, gender-based violence response, and adolescent education. We accompany girls, young women and key populations across Kisumu County and push for county-level policy and budget changes that make their rights real.

Contact
hello@afyarightsinitiative.or.ke
+254 717 558 070
Kisumu, Kenya
Document
ARI-RS-2025-04
v1.0
August 2025
Registration
Community-based organisation registration with the State Department for Social Protection, Kenya.
License
Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)
Suggested citation

Afya Rights Initiative (2025). Girls' Retention Study, Nyando 2025. Kisumu, Kenya. ARI-RS-2025-04. Released under CC BY-NC 4.0.

Back to all documentsRequest data or partner with us →