Scaling teacher training In Sierra Leone: how can cascading models deliver results?

• In Event: Teacher education, pedagogy, and professional development

Proposal

Scaling teacher training in Sierra Leone: how can cascading models deliver results?

Context
The goal of this presentation is to examine a cascade training strategy in the context of large-scale training for primary school mathematics teachers in Sierra Leone. We will discuss the specific features of the cascade model that helped maximize its effectiveness on teachers, while also addressing the limitations of this approach and exploring possible solutions to enhance its impact.

A study was conducted to evaluate the effectiveness of a nationwide numeracy training initiative for Primary 1 (P1) teachers in Sierra Leone, launched by the Ministry of Basic and Senior Secondary Education (MBSSE) with support from UNICEF and the Global Partnership for Education (GPE). As part of Sierra Leone’s 2023–2027 foundational learning strategy, the cascade training strategy aimed to build the capacity of teachers, school leaders, and the government to implement effective teaching practices, such as structured lessons and the Concrete–Practical–Abstract (CPA) approach, aimed at enhancing conceptual understanding through staged, manipulative-based learning. Specialized teaching and learning materials (TLMs) were also developed, including a numeracy workbook and a teacher guide. The training has been delivered in clusters (in each of the 16 districts in Sierra Leone) following the 3-stage cascading process where an expert trained 30 master trainers who in turn trained 300 trainers who ultimately trained 12,500 grade 1 teachers. This study focuses on the last stage of this process, in which 300 trainers trained all the grade 1 teachers of Sierra Leone alongside their school leaders.

This model was selected for its proven impact in low-resource settings and its alignment with national priorities to improve instructional quality and learning outcomes in early primary grades.
To assess its effectiveness, the study employed a mixed-methods design. A standardized pre- and post-training test—covering key training modules such as formative assessment, lesson planning, and use of the new TLMs—was administered to over 7,500 participants across 16 districts. Results were complemented by more than 5,600 qualitative feedback forms that shed light on teachers’ perceptions of training content, quality, and delivery, as well as observations to assess fidelity of training delivery across venues.
The training process for 12,500 teachers
The teacher training process lasted seven consecutive days. Teachers had access to training handbooks and excerpts from the teacher guide and P1 workbook.
1. The training began with three days of theory focused on the specific features of the teaching-learning materials distributed to teachers.
2. Teachers were then trained on the structure of the lesson plans proposed in the teacher guide and on how to deliver lessons.
3. Then, three days were dedicated to practical simulation of lesson plans by teachers.
4. A final day was devoted to teachers constructing lesson plans in connection with activities in the workbook.

The challenge of this training was to achieve maximum impact among all grade 1 teachers in Sierra Leone through a cascade training program for which the limitations are well known.
The main weakness lies in the risk of dilution of content at each level of transmission, leading to a loss of the initial message or misinterpretations (Hayes, 2000; Bett, 2016). This weakness is exacerbated by other factors cited by Bett (2016), such as: 1- failure to take into account the needs of teachers, 2- failure to take into account the local context, 3- insufficient skill levels among secondary trainers.
However, the cascade model is not inherently flawed, but is often poorly implemented (Hayes, 2000). Hayes’ work highlights five key principles for maximizing the impact of cascade training:
1. The method of conducting the training must be experiential and reflective rather than transmissive.
2. The training must be open to reinterpretation; rigid adherence to prescribed ways of working is unlikely to yield results.
3. Expertise must be diffused through the system as widely as possible, not concentrated at the top.
4. A cross-section of stakeholders must be involved in the preparation of training materials.
5. Decentralization of responsibilities within the cascade structure is desirable.


Embedding the five key principles into the training in Sierra Leone
To maximize the effectiveness of our cascade training approach and minimize its limitations, we integrated the principles proposed by Hayes (2000) throughout delivery, while incorporating additional strategies that we believed would enhance its impact:
a) The training was based on instructional materials designed specifically for P1 teachers, including a student workbook and a teacher guide.
b) Teacher trainers were provided with detailed training documents and step-by-step guidelines to minimize content dilution.
c) Training sessions were held within each district, close to the schools of the participating teachers.
d) The training content focused on the TLMs that teachers would be using in the classroom.
e) Discussion sessions were organized after each activity to encourage active participation and reflection among trainees.
f) The focus of the training was on practical teaching strategies and lesson delivery, with teachers encouraged to integrate the theoretical principles covered during the training into sample lesson plans.
g) Trainers were trained to provide constructive feedback to teachers.

Some results
The training significantly improved teachers’ knowledge of numeracy instruction, with gains observed across gender, roles, and regions. Female teachers slightly outperformed males. School leaders and government officers scored relatively higher than regular teachers, highlighting the value of school-based coaching. Participants achieved stronger results in the CPA and lesson planning modules, while scores on the formative assessment and learning outcomes modules remained comparatively lower. Feedback stressed the need for more hands-on practice, classroom simulations, and access to the full teaching materials designed for use in classrooms. Effective strategies such as collaborative planning, modeling, and using local resources enhanced engagement and learning. These findings offer valuable insights for strengthening and scaling teacher training in resource-limited contexts.
Conclusion
During our presentation, we will present in greater detail the results of our analyses. Based on those results, we will also provide a series of recommendations to strengthen teacher capacity and discuss strategies for a continuous professional development process grounded in the principles outlined by Cordingley et al. (2023).

Authors

• Stephane Cyr, Université du Québec à Montréal, UQAM

• Marco Valenza

• Thomas Dreesen, UNICEF