Differentiated instruction in multigrade preprimary classrooms in Kenya

Differentiated instruction, “the process of modifying or adapting the curriculum according to the different ability levels of the students in one class” (UNESCO, 2004, p. 14), widely has proved to be an essential component of high-quality instruction that supports student learning. All children develop at their own pace and thus have different needs when learning new content (Tomlinson, 2000; UNESCO 2004; Shillady, 2013; Slavin, 1987; Subban, 2006). Reviews of various intervention programs at scale point to the importance of aligning instruction to the goals and needs of students (Hwa, Kaffenberger & Silberstein, 2020). In many ways, this alignment is seen as a key component of high-quality instruction and can be used with all subject areas, ages, and instructional models (Mathes et al., 2005).

Differentiation is also widely acknowledged to be difficult to use in the classroom. It can be applied in multiple aspects of instruction, such as ability grouping (e.g., teaching children by assessed skill levels), formative assessment to plan instruction, and adaptive teaching to adjust instructional strategies during lessons (Mulroy & Eddinger, 2003; Parsons et al., 2013; Slavin, 1987). The difficulty of applying differentiation is especially true for teachers in low- and middle-income countries (LMICs) where low levels of teacher education and low availability of high-quality materials have been exacerbated by the pandemic (Kett et al., 2018; World Bank, 2022). In fact, although many recommendations for quality instruction call for teachers to differentiate instruction, there is little rigorous evidence of existing practices in differentiation in LMICs. In this paper, we assert that if a government desires to introduce and sustain differentiation as a widespread practice, we must learn from what teachers are already doing and build from there.

Deficit approaches have been widely used to characterize education in LMICs around the world (Aikman et al., 2016). Seen through a deficit lens, teachers lack certain skills and knowledge needed for successful student learning to occur. Often, programs seeking to improve the quality of education deliver trainings and classroom materials based on research and best practices from high-income contexts, instead of documenting and incorporating existing expertise (Piper et al., 2018). This study provides a counternarrative to this deficit approach with our underlying belief that teachers have skills and knowledge that are worth uncovering. These existing skills and knowledge can allow for approaches that build on teachers’ strengths and are effective and contextually relevant.

We sought to uncover natural teacher practices associated with differentiated instruction in preprimary classrooms in Kenya. We used classroom observations and teacher interviews to capture the ways in which teachers differentiate instruction, as well as teacher views on why they differentiated instruction. To do this, we took advantage of a unique opportunity within the Tayari Program, a Children’s Investment Fund Foundation (CIFF)–funded program implemented in preprimary classrooms in four counties in Kenya. In particular, we focused on multigrade classrooms (which combined children between the ages of 3 and 6 years old) as we theorized that differentiated instruction would be more likely to occur within multigrade classrooms than in single-grade classrooms.

Whereas we acknowledge that differentiated instruction involves more than differentiation by grade levels, we nevertheless assert that multigrade classrooms where differentiation might occur by grade levels represent an entry point into the practice. We also acknowledge that the Tayari case study was a special case, as teachers were provided with high-quality materials and training, and we used a purposeful sample to select teachers who were identified by program staff as the most likely to differentiate instruction. We did this because we were interested in identifying the beginnings of a differentiation practice that could be modified and adapted—to single-age classrooms, classrooms without high-quality materials, and classrooms with many children—but would still have its roots in Kenyan classroom practice. To do this, we used a case study methodology (Yin, 2017) that allowed us to examine a complex, real-life event (i.e., classroom teaching) to better understand how teachers make decisions about differentiating instruction in classrooms.

Through 16 observations in eight classrooms, we uncovered two ways in which teachers differentiated instruction. We then used data from our teacher interviews to ask teachers about certain parts of the lesson we observed and why they made certain choices around differentiation. From our findings, we made conclusions about what types of support teachers may need to effectively differentiate instruction in classrooms in contexts similar to that of Kenya.

The case study described in this paper took place within the Tayari Early Childhood Program in Kenya (2016–2019). The larger aim of this program was to pilot and test the effectiveness of a high-quality early childhood education model, which promoted school readiness outcomes for children through the development and provision of teacher guides and student books for language, mathematics, social skills, and life skills, and through ongoing teacher support and training. Given the focus of our study on differentiated instruction, we turned our attention toward the multigrade schools within the Tayari program to better understand how teachers work around barriers to differentiation. The Tayari program did not provide teachers any guidance on how to differentiate; in multigrade classrooms, teachers were only provided with materials of the highest-grade level, which offered an opportunity to uncover teacher-initiated practices in differentiation.

The barriers in this and similar contexts include large class sizes, lack of systematic training for teachers in primary and preprimary, scarcity of resources, low attendance rates, and recently, school closures due to the pandemic. These barriers make it difficult for teachers to identify and instruct at different ability levels in the classroom. Perhaps because of these difficulties, many recommendations for high-quality instruction in LMICs refer to differentiated instruction as essential but fail to provide concrete recommendations to support teachers to differentiate (Sitabkhan & Platas, 2018).

Many researchers within high-income countries recommend providing differentiated instruction, particularly for literacy instruction and for children learning in their non-native language (Hong et al., 2012; Taylor et al., 2000; Cheung & Slavin, 2012). Most research from LMICs focuses identifying learner needs through summative and formative assessment. Studies by Stern et al. (2018) in Indonesia and Owusu (2016) in Ghana describe pre-assessments as a method to identify learner needs. While informative as to how first steps toward differentiation can occur in LMICs, these studies do not address how teachers within a classroom can meet the needs of diverse learners.

A notable exception to this is Msimanga (2020), who provides concrete recommendations of how teachers can differentiate instruction in multigrade primary classrooms in South Africa. The authors used interviews to better understand the nature of instruction in nine classrooms. Specifically, the author found that teachers were likely to group students by grade. Teachers taught each grade group separately when topics in the syllabus were different. When topics were similar, teachers would instruct all the grades at the same time. When teaching all grades together, teachers shared different strategies they implemented, including using peer tutors, implementing small group instruction, and creating separate work for each grade level.

Another body of related research has described and evaluated interventions in which children who needed extra support were provided instruction outside of regular schooling (Gove et al. 2017; Banerjee et al., 2007). One technique gaining global recognition is Teaching at the Right Level (TaRL), developed by Pratham in India (Banerji & Chavan, 2020). TaRL uses regular assessments to group students by ability level and offer targeted support. For example, if students are struggling with letters, they are grouped with other similar-leveled students and receive instruction targeted to learning letters and sounds. This instruction could be during school or after school. Although they indicate how first steps toward differentiation can occur, these studies do not address how teachers within a standard classroom can meet the needs of all learners, not just struggling learners, at the same time. TaRL, for example, can require extra teachers or facilitators to meet the needs of different groups of students.

In Peru, an intervention in preschool mathematics sought to do just this. The intervention supported teachers to use a simple method to ascertain the level of students in certain core concepts and then target instruction appropriately through the use of 45-minute sessions per week, as well as support from a visiting teacher assistant (Gallego, et al., 2017). However, we did not find research on differentiated instruction at the preprimary level in Sub-Saharan Africa. Thus, our study fills a gap in describing existing practices in detail by capitalizing on this case study of multigrade classrooms within the Tayari Program.

The research questions our study addresses are:

• How did teachers in multigrade preprimary classrooms in Kenya differentiate instruction during language and mathematics lessons?

• What types of differentiation did teachers use?

We found that all eight teachers we observed used differentiated instruction in some way, although the degree of differentiation varied. We organize this section around descriptive results detailing two aspects of this differentiation and the evidence around them from both the classroom observations and the teacher interviews. Throughout these sections, we present findings from across both observed age-group configurations (Baby/Preprimary 1/Preprimary 2 [3–6-year-olds] and Preprimary 1/Preprimary 2 [4–6-year-olds]). Following the results, we describe in the discussion section how these results allowed us to address each of the research questions.

The results of this study show that the teachers we observed in Kenya used differentiated instruction during reading and math lessons in preprimary multigrade classrooms, despite challenges such as small physical classrooms and no prior training in differentiating instruction. We discuss the how the data address the research question, and then focus on how the findings shed light on the effectiveness of the differentiated instruction.

There are limitations to this study. The sample size was small and thus does not allow for generalization to the population of multigrade teachers in Kenya. In addition, the teachers were chosen because they had demonstrated effective techniques for multigrade teaching as reported by supervisors. Our purposeful sample was drawn to align with case study methodology, which calls for studying a real-life event (successful teachers differentiating instruction), making this population of teachers even less representative of other multigrade preprimary teachers in Kenya. However, the purpose of this study was not to understand what typical multigrade teachers do but instead to examine a case study of exemplar teachers to uncover and detail productive practices, understand ongoing challenges, and suggest productive next steps for supporting preprimary teachers in Kenya. Future studies should include a wider range of multigrade teachers in Kenya to better understand whether the practices found in this exploratory study are common and how they may differ when teachers are not high performing.

Based on these findings, we present several recommendations for supporting more teachers to use differentiated instruction through both professional development and classroom materials. Importantly, we believe that our findings reveal the presence of the roots of a complicated practice that are encouraging and confirm our hypothesis that teachers are making efforts to differentiating instruction in difficult contexts.

First, our results show that almost all differentiation occurred according to assigned grade levels (Baby, Preprimary 1, and Preprimary 2). Although high-quality differentiation would ideally target instruction based on the child’s ability rather than the child’s age/grade level, we find grade-level differentiation to be a positive first step toward incorporating differentiated instruction in the classroom. As with any attempt to change teacher practice, supporting teachers to differentiate instruction by ability is a process that takes time. It may be that using assigned grade levels is a first step to encourage teachers, and that the pre-determined groupings allow teachers to be comfortable with the idea of supporting different groups of students doing different activities at the same time within the same lesson. This is especially important given the large class sizes and lack of resources in most classrooms. In addition, recent research in Tanzania (Jukes, Sitabkhan, & Tibenda, 2021) investigating social and cultural influences on instructional decisions found that teachers valued togetherness and fairness in their teaching. For some teachers, targeting instruction at students’ achievement levels was seen as unfair and could undermine the class’s sense of togetherness. Grade-level differentiation could be a first step to overcoming this perception.

Second, curriculum developers should be aware of the ways that materials, including teacher guides, may be used in multigrade settings. Within the Tayari classrooms, for example, we found that providing teachers with only the teacher’s guide for the highest-grade level within their classroom was problematic. Teachers tended to teach the lesson as is, without much differentiation in questioning, during the whole group lesson. Many of the individual/small group activities that were differentiated were not aligned with the whole group lesson and objective, possibly causing a disconnect for younger students. For example, a whole group lesson was about putting together objects, a foundation for addition. In small groups, younger students were asked to model numbers with plasticine, which is cognitively a very different task for students than putting together objects. Therefore, curriculum developers should consider the needs of multigrade teachers and create materials accordingly. For example, whole group lessons could include questions at two or three different levels, providing guidance on harder and easier questions. During small group activities and/or individual work time, materials could provide guidance on two or three levels of activities aligned with the same objective. For example, taking the putting together example from earlier, the number of objects being put together can vary according to the age/level of the child, but the overall objective remains the same.

Third, teacher trainings can address the perceptions that teachers have about on whom to focus the most attention. In our findings, we saw variation in this. Some teachers tended to focus on the oldest students, as they were preparing for Grade 1 entry. Other teachers tended to focus on the younger students, as they needed the most support. Neither is wrong or right. Teacher trainings can help teachers understand how to vary support according to the activity and needs of the students.

This study attempted to disrupt the deficit model that is often behind many early childhood interventions in LMICs. Instead of assuming that teachers do not know how to differentiate instruction, and developing an intervention based on research from high-income countries using differentiated instruction, we sought to reveal existing practices that teachers are already using. We documented several practices that, while not perfect, were encouraging first steps toward a complicated practice. We hope that future research can document more of these existing practices, which can be modified to address the challenges teachers face and expanded to other teachers in similar circumstances.

We would like to thank the Tayari Early Childhood teachers that allowed us to observe their classrooms and shared their view on instruction with us.

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

RTI Press Associate Editor: Rebecca Flueckiger