Recontextualization of learner-centred pedagogy in Rwanda: A comparative analysis of primary and secondary schools

ABSTRACT

Learner-centred pedagogy (LCP) has become a global pedagogy and has been adopted in sub-Saharan African countries such as Rwanda, despite ample evidence of implementation failure. Most research has examined its implementation at either the primary or the secondary level. However, this qualitative study adopts a comparative approach and seeks to explore how Rwandan primary and secondary school teachers define, perceive and recontextualize LCP. The study is based on interviews and classroom observations of 12 effective primary and 12 effective secondary teachers working in eight well-performing schools; the analysis draws on Schweisfurth’s minimum standards for LCP. This Rwandan case reveals that the majority of primary and secondary teachers stimulated open and respectful classroom interactions. However, recontextualization of constructivism differed substantially between the two groups. Furthermore, the findings highlight the importance of more research into the dynamics between the various standards for LCP, and the interaction of these standards with contextual factors.

KEYWORDS:

• Learner-centred pedagogy
• rwanda
• recontextualization
• education reform
• teachers

Introduction

In 2016, Rwanda adopted learner-centred pedagogy (LCP) as its official pedagogy, together with competency-based education, for pre-primary, primary and secondary education. By that point, LCP had become a global pedagogy and had been adopted in various countries in sub-Saharan Africa (SSA) (Vavrus, Thomas, and Bartlett 2011; Schweisfurth 2011). Many developing countries embraced LCP because they perceived this pedagogy to be a solution to a myriad of societal and educational problems, such as social inequality in South Africa (Nykiel-Herbert 2004) or poor student performance in Uganda (Altinyelken 2010). Rwanda was no exception to that rule. The new curriculum with LCP as the prescribed pedagogy was not only expected to improve the quality of education, but also to equip ‘a critical mass of the population with knowledge, skills and attitude to be highly competitive in the global market’ (REB 2015, 1). One difference from other SSA countries might be the imperative for educational reform. The genocide of 1994 left the country and the education system in ruins. In order to prevent another war, the government’s first concern was the creation of an inclusive education system offering equal educational opportunities to the two main ethnic groups, Hutus and Tutsis. With the help of the international community, Rwanda achieved impressive results regarding school enrolment; overall attendance in primary schools increased from 62.5% of eligible students in 1990 to 97.7% in 2016 (Abbott, Sapsford, and Rwirahira 2015; MINEDUC 2017). However, this success further jeopardised the yet poor quality of education (USAID 2014; ICAI 2012).

The government perceived the new curriculum together with LCP as the means to improve educational quality, despite ample empirical evidence about the disappointing results in other SSA countries (Tabulawa 2013; Guthrie 2018; Harber 2017; Schweisfurth 2013). In fact, the implementation of LCP in developing countries is ‘riddled with stories of failure grand and small’ (Schweisfurth 2011, 425). There are three main reasons for this failure. First, there is the nature of the reform itself; it is often too hasty, complicated and in a language incomprehensible for teachers (Schweisfurth 2013). Second, in most developing countries there is a lack of resources, such as textbooks and other teaching materials (Vavrus, Thomas, and Bartlett 2011). Third, social-cultural beliefs in Africa about classroom relations, knowledge and knowledge construction are perceived to be incompatible with LCP (Tabulawa 2013; Guthrie 2018).

The empirical evidence to date has led to contrasting conclusions and opposing positions. On the one side, there are scholars who have argued that, due to the above-mentioned incompatibility, developing countries should not opt for LCP at all (Guthrie 2018; Tabulawa 2013). Instead, it would be better to revert to the more traditional, teacher-centred teaching style and to work from there. On the other side, there are scholars who, despite the many hindrances in implementation, value the potential of LCP (Schweisfurth 2013; O’Sullivan 2001; Tikly 2019). They have pleaded for local adaptations or transformations that provide for acknowledgement of context and the wider system of which education is a part. Harber’s (2017) recent work on schooling in sub-Saharan Africa pointed towards multiple forms of hybrid pedagogies and promising exceptions that contradict the ‘story of failure’. The hybrid forms and the exceptions stem mainly from the past decade, while earlier studies often emphasised failure.

This study contributes to this ongoing discussion by adding empirical evidence about how Rwandan primary and secondary teachers recontextualize LCP to adapt it to their local contexts and thus to the dynamics between the global and the local. This investigation has two research questions: 1) How do effective Rwandan primary and secondary teachers define and perceive LCP? and 2) How do these teachers recontextualize LCP in their classrooms? This research also addresses an empirical gap in current literature concerning LCP in Rwanda. Furthermore, the majority of studies have focused on a single level of education, mostly primary, whilst in this study the perspectives and experiences of primary and secondary teachers are compared. This comparison allows for a more in-depth analysis of the recontextualization process.

Learner-centred pedagogy

Learner-centred pedagogy is presented in a plethora of terms, such as student- or child-centred pedagogy, active learning, inquiry-based learning or progressive education. All these terms have in common that they arise from the learning theory of constructivism, which is grounded on the work of Vygotsky and Piaget (Vavrus, Thomas, and Bartlett 2011). Compared to traditional teaching, where the teachers’ role is to deliver undisputable knowledge to the learners, LCP assumes that knowledge is created through interaction and that it builds upon previous knowledge and personal experiences. It is the teachers’ responsibility to enhance this interaction and to stimulate knowledge construction that connects to learners’ next zone of development (Schuh 2003).

Due to the plethora of terms and the various interpretations, LCP is not easy to define. To deal with this challenge, yet still allow for cultural variants, Schweisfurth (2013) defined seven minimum standards. Each standard adheres to a basic principle of LCP and is derived from empirical evidence from LCP implementation. The standards are: 1) lessons need to be engaging to students and motivating them to learn; 2) there is mutual respect between teachers and learners; 3) what is taught builds on learners’ existing knowledge and skills; 4) dialogue is used in a variety of forms, not only for transmission; 5) curriculum is relevant to learners’ lives in a language that is accessible to them; 6) curriculum and pedagogy contain skills and attitude as well as content, and include critical and creative thinking; 7) assessment is meaningful and supports the learning process. The set of standards serve as a culturally – or contextually – adaptive framework. For instance, how students are being motivated and engaged is open to local adaptation, as is the amount of dialogue used or the extent to which relationships between teachers and students are formal.

Due to their adaptive character, these standards were chosen to analyse the recontextualization process in the case of Rwanda. This study thereby concurs with the in-depth classroom analysis of two Kenyan teachers conducted by Lattimer (2015), who also engaged with Schweisfurth’s standards. The Kenyan study demonstrated that these standards contributed to a more profound understanding of LCP in the local context and that rich examples helped to build connections between policy and practice.

Convergence, divergence and recontextualization of LCP

The globalisation of LCP suggests a convergence of pedagogy, whilst the empirical evidence rather points towards divergence. Convergence is explained by the unquestioning belief in the salutary effects of LCP and its image as ‘best practice’ (Anderson-Levitt 2008; Carney, Rappleye, and Silova 2012). The World Bank and OECD have reinforced the quest for ‘what works’ and ‘best practice’ by relying on and referring to quantitative evidence-based data from large-scale international studies, such as PISA and TIMMS. This trend has criticised by various comparativists during the past decade (e.g., Klees et al. 2020; Edwards 2020; Auld and Morris 2014). These scholars critiqued the way these international organisations present the findings from such studies: they present correlation as causation, are selective in the evidence to support their ideology and fail to acknowledge context and the many interconnected variables that affect learning outcomes.

Pedagogy is also subject to this trend, and although the ample qualitative evidence points towards a wide range of interpretations and applications of LCP, there are also commonalities. First, based on empirical and theoretical analyses, there seems to be universal support for some pedagogical components, such as mutual respect between teachers and learners, and a safe environment and classroom interactions (Schweisfurth 2013). Second, research on the implementation of LCP in developing countries revealed a tendency to embrace similar methods, such as group work and the use of teaching aids (Harber 2017).

Divergence is explained by the active role and agency of actors in the recipient countries and the power they use to adapt a borrowed pedagogy to their local context. In this study, the term recontextualization is used for this process, but this term is interchangeable with ‘indigenisation’, ‘modification’, ‘translation’, ‘transformation’, etc. In short, recontextualization considers how ideas, policies, or in this case, pedagogies, are transformed as they move among fields, actors and contexts. During this process, pedagogies take on new meanings and interpretations (Alexander 2001; Crossley 2019).

There are various ways in which teachers react. The most extreme reaction is ‘tissue rejection’: when context and agency of teachers are not acknowledged, teachers might reject the reform in their actual classroom teaching (Harley et al. 2000). Another response is to apply LCP only superficially or in a simplified manner. For instance, Namibian teachers only applied certain communication strategies after the strategies had been simplified with prefab templates, although that led to a limited and less flexible use of those strategies (O’Sullivan 2001). A third reaction is the creation of a mixed or hybrid pedagogy: a combination of traditional and more progressive approaches. For instance, in China, teachers mixed Confucianism and learner-centred practices (Tan and Chua 2015). Lastly, the adoption of LCP can also result in unforeseen or even counterproductive results when the context is not conducive to supporting this pedagogy. For instance, in Turkey, research assignments were introduced as a tool to improve students’ skills at retrieving and selecting relevant information, but lack of internet access and reference books in many households compelled many students to farm out their assignments to others (Altinyelken 2013).

It is important to note that although various scholars have acknowledged the divergent ways that LCP can appear due to contextual variations (O’Sullivan 2004; Thompson 2013; Schweisfurth 2013), empirical evidence on its implementation reveals certain similarities. Besides the earlier indicated reasons for imperfect or failed implementation, there are four other hindering factors that have emerged repeatedly from case studies. First, there is the examination system, which often still relies on memorisation despite the curriculum reform. Furthermore, most SSA countries have a foreign language of instruction (LoI), often the one of the former coloniser. A foreign LoI seriously hinders interaction, key in LCP, in classrooms. Third, like their teachers, students must adapt to this new approach, and often they resist the new active role that is expected from them. Lastly, many reforms have failed due to low teacher morale (Altinyelken 2010; Tabulawa 2013; Schweisfurth 2013). Nonetheless, there is a consensus about the key role of context (Cowen 2009; Phillips and Ochs 2003); in the next section, the contextual factors that characterise the Rwandan case are outlined.

The Rwandan context

Rwanda is a late adopter of LCP, and despite ample evidence of implementation failure, key stakeholders did not question the suitability of this pedagogy. This lack of criticism can be explained by the various global influences, such as the aforementioned quest for best practices and the omnipresence of international aid agencies, but also by South to South policy transfers (Van de Kuilen et al. 2019). As a member of the East African Community, the country had to adhere to the regional curriculum framework that included LCP.

The implementation of LCP started in January of 2016 without an official pilot. Instead, the Government of Rwanda (GoR), with the Rwandan Education Board (REB) as the executive organisation, used a cascade training model that prepared schools and teachers for the new curriculum and pedagogy. The cascade model included trainers at national (national trainer, NT), district (district master trainer) and school level (school subject leader, SSL). NTs received a seven- to 10-day training from REB three times, on the following topics: 1) general introduction to the new curriculum; 2) scheme of work and lesson plans; 3) assessment. Those teachers appointed as SSLs received a one-week training from the district trainers, who in their turn were mostly trained for 10 days by the NTs. In-service training was the most vital component in this cascade model, and consisted of standardised and non-standardised in-service trainings. The standardised trainings were prescribed by REB, while the non-standardised trainings that followed after allowed for school-specific demands (MINEDUC and REB 2015). However, most teachers had to start working with the new curriculum with hardly any preparation, since the trainings coincided with the actual implementation.

Two other factors complicated the implementation. First, as in many sub-Saharan countries, Rwandan teachers face a lack of resources, including time. As a response to the high rise in enrolment, primary teachers had to work double shifts that left them with little or no time for lesson preparation. In 2018, the GoR ordered a gradual phase-out of double shifts, but to date, some teachers still have to work double hours. Regarding resources, the GoR decided to write and publish new textbooks that would match the learners’ socio-cultural environment. However, the execution of this lagged behind, and resulted in a situation where teachers had to start working with the new curriculum with hardly any accompanying textbooks (RENCP 2019).

Second, in 2009 the GoR unexpectedly and swiftly switched the LoI for upper primary and secondary education from French to English. Teachers had hardly any time to respond, adding an extra challenge to their work (Abbott, Sapsford, and Rwirahira 2015). The use of a foreign LoI in Africa is often discussed in relation to learning outcomes. In brief, opponents have argued that a foreign LoI negatively effects learning outcomes, whilst proponents perceive English or French as a prerequisite for social mobility (Brock-Utne 2012; Trudell 2016). A decade after the introduction of English, the GoR has acknowledged low English proficiency and continues with interventions to improve teachers’ language skills (RENCP 2019).

Method

Sample

The sample in this study comprises 12 primary and 12 secondary teachers from four primary and four secondary schools. Access to schools was obtained through a research permit from the Ministry of Education (MoE). The schools were purposively selected with the help of a school inspector from the MoE on the criterion of being a high-performing school. The rationale behind this selection is grounded in the chosen mutual adaptation perspective, reasoning that more insight into recontextualization is gained from effective rather than under-performing teachers. The status of being a high-performing school is mainly based on examination results, which – as acknowledged by the researchers – is an imperfect proxy for good application of LCP. The secondary schools in this sample are all boarding schools. Contrary to non-boarding public schools, these schools have a long history and were, in the past, meant for the elite (Williams 2019). Currently, students with high examination scores are invited to boarding schools. This explains probably why, for secondary education, only boarding schools were selected. Most students from the selected schools come from middle class families. The primary schools in this sample also offer public basic secondary education, but since only the primary department of those schools performed well, the secondary teachers from these schools were not included in the sample. The student population of the primary schools is mixed; in rural schools, the majority of the children are poor, while in the capital the socioeconomic status of the parents varies from poor to lower middle class. Middle class children often attend private primary schools. Lastly, the sample consists of a mix of rural and urban schools to ascertain if the geographic location was of influence.(Table 1)

Table 1. School sample background information

School	Location	Level	Boarding	Average class size of observed lessons
1	Rural	Primary	No	51
2	Urban	Primary	No	44
3	Rural	Primary	No	65
4	Urban	Primary	No	43
5	Urban	Secondary	Yes	44
6	Rural	Secondary	Yes	28
7	Urban	Secondary	Yes	38
8	Rural	Secondary	Yes	49

Three teachers from each school were selected with the help of either the principal or deputy school leader. The main criterion was their status of being an effective teacher and being perceived as a ‘good implementer of LCP’. All had the required qualifications to teach at their level. Among the primary teachers, three had been appointed as a national trainer (NT) or school subject leader (SSL). In the sample of the secondary teachers, five had been appointed as NT or SSL. The other teachers received, to a varying extent, in-service training from their colleagues or support from non-profit organisations prior to the new curriculum. In general, the cascade training model, as described in the previous section, led to substantial dilution; whereas national trainers could count on between 21  to 30 days of training, other teachers in this sample had to rely on only a few in-service sessions of two to three hours.(Table 2)

Table 2. Teacher sample

	Gender	Age	Educational level	Years of experience	No. of teachers appointed as trainer
Primary teachers (N = 12)	Female: 6 Male: 6	20–29: 5 30–39: 3 40–49: 3 50–59: 1	A2 level (upper secondary): 12	1–5: 3 6–10: 6 11–15: 0 16–20: 2  > 20: 1	2 national trainers (NTs) 1 school subject leader (SSL)
Secondary teachers (N = 12)	Female: 2 Male: 10	30–39: 9 40–49: 3	A0 (bachelor): 12	0–5: 1 6–10: 8 11–15: 0 16–20: 0  > 20: 3	2 NTs 3 SSLs

Data collection

Data collection took place from September, 2017 to March, 2018. The first author, who conducted the field study, informed all teachers about the purpose of the study. In general, there was great willingness to participate. Although none of the teachers explicitly asked for it, all were given pseudonyms. Each teacher was observed and – in all but one case – afterwards interviewed, reasoning that this order would give more opportunity for the researcher to ask questions about intentions behind observed teaching practices. The presence of an observer might lead to a change in teacher behaviour. Considering the observed student actions and reactions, this was either very mild or not at all the case.

The lessons lasted 40 minutes in primary school and 50 minutes in secondary. The average class size in primary school was 51 children, with a minimum of 38 and a maximum of 72. In secondary classes, the number of students ranged between 20 and 52, with an average of 40. Male teachers outnumbered female teachers in the secondary school sample. The lessons were observed using an observation form that was derived from the minimum standards for LCP (Schweisfurth 2013).

The semi-structured interviews were conducted whenever possible on the same day as the observed lesson. The questions focused on four main topics: 1) training and support before and during the implementation; 2) implementation of LCP, including the experienced challenges; 3) expectations; and 4) opinions about LCP. The teachers could choose to do the interview in French or English. Four primary and one secondary teacher preferred to do the interview in French. The interviews were conducted in a quiet and private space to allow teachers to speak freely. All but four interviews were audio-recorded and afterwards transcribed. The researcher took detailed notes during the four other interviews. The average duration of the interviews was 40 minutes.

Data analysis

All interviews were transcribed verbatim and read multiple times to detect patterns and emerging themes. Then, data were coded using a software programme (Atlas-ti), based on a list of codes derived from literature. After coding a first set of interviews, codes were discussed with another member of the research team and further honed in line with the emerging themes. The revised code list was used to analyse all interviews. The main codes included the following: definition, opinion, challenges, expectations, teaching and learning methods, knowledge construction, student and teacher role, and language.

The lesson observations were carried out with reference to the minimum standards. Some of the standards are indisputable, such as the use of dialogue other than transmission (standard 3), whilst others required additional operationalisation. For instance, motivation and engagement, standard 1, was operationalised by the percentage of learners (> 70%) that was active throughout the lesson. Atmosphere and conduct, standard 2, was understood as good classroom management, that is, all learners listened to the teachers and their fellow students and there was no disturbance of the learning process. Lastly, critical thinking (standard 6) was defined as higher-order thinking, where learners were challenged to analyse, evaluate or create knowledge. The authors acknowledge that this operationalisation already accounts for a certain degree of contextualisation and that it reflects a researcher effect.

During the time of data collection, the first author was a resident of Rwanda and worked as an education consultant in various roles and education sectors. She was involved in some projects that supported the implementation of LCP, but not in the schools included in this study. She made numerous visits to a wide variety of schools and met the main actors involved in the dissemination of LCP. These experiences helped her to get an in-depth overview of the educational landscape in Rwanda. This contextual knowledge also facilitated the contact and conversations with teachers. However, she conducted this study as an independent researcher, with the University of Rwanda as the affiliated institution and with permission of the Ministry of Education.

Findings and discussion

This section begins with a description of how effective Rwandan primary and secondary teachers define and perceive LCP (research question 1). Their perceptions include their opinions and expectations and the challenges they experience. These factors were included because teachers’ beliefs as well as the contextual factors are important with regard to the extent to which they are willing and able to comply with the curriculum reform (Biesta, Priestley, and Robinson 2015). That is followed by a description of actual classroom recontextualization (research question 2), starting with an overview of the standards that were mostly observed, and complemented with a description of the way teachers interpreted these standards. These interpretations are described on the basis of three themes that emerged from the interviews and observations, namely constructivism, classroom relations and language.

Teachers’ definition and perceptions of LCP

Both primary and secondary teachers defined LCP as a teaching approach where learners are more active than the teacher, and where the teacher has to act as a facilitator. The main difference between primary and secondary teachers concerned knowledge construction. Whereas primary teachers expected that learners are primarily responsible for knowledge construction, secondary teachers perceived it as their responsibility to evoke knowledge construction through questioning, judging what is right or wrong, or adding to what learners bring into the classroom. Most definitions by primary school teachers resembled the following:

It is a lesson focused only on children, not on me. The children talk more and work more than me. It is a method that attracts learners to work. (Adolphe, male, primary school 2)

All teachers were positive about LCP for several reasons. The first reason is because of the high expectations they have for students and society. Due to this new pedagogy, they foresee that their students will learn how to collaborate and live together, which is not only beneficial for themselves, but also for society, since that may prevent another war:

They [previous generation] participated in the genocide because they didn’t know how to live together. But with this methodology, they learn to live together. (Odille, female, primary, school 2)

Second, teachers expected that due to LCP, students will learn how to apply knowledge, which will not only give them better preparation for the labour market, but will also boost the economy. Third, teachers observed that with LCP, students are more active and motivated, and they expected that it will result in better learning outcomes.

When explicitly asked for disadvantages, other than challenges, only a few teachers mentioned shortcomings like noise or that not all students can participate in groupwork. Regarding resistance, the opinions varied. One primary and three secondary teachers mentioned resistance among their colleagues, but at the same time there was an unconditional acceptance of the new curriculum.

The curriculum is provided by the government. We do it, even without all the materials. There is no way to go against something that is designed by the government. We are implementers of what the government wants us to do. (Jean Claude, male, secondary, school 7)

The positive view of LCP was tempered once challenges were being discussed. The main – often mutually reinforcing – challenges were lack of resources, lack of time and classroom size. Nearly all teachers mentioned the absence of (sufficient) books. The challenge mentioned next most was time. Nearly all teachers perceived LCP as more time-consuming than traditional teaching, mainly due to the increased classroom interactions and the realisation that LCP requires more lesson preparation.

Classroom recontextualization

The recontextualization of LCP with reference to the minimum standards is listed in Table 4. This table shows that, contrary to definitions and perceptions, the actual implementation of LCP varied between and within groups, except for the overall use of dialogue (standard 4). The differences concerned the standards that teachers addressed; secondary teachers, on average, addressed more and other standards than their colleagues in primary classrooms. In primary education, teachers demonstrated, besides standard 4, motivation and engagement of learners (standard 1) and relevant topics, in a language accessible to the children (standard 5). The latter is remarkable, considering the fact that in upper primary learners are being taught in English, a foreign LoI. However, most teachers, being aware of this hindrance, demonstrated language support in various ways. Additionally, the content of the new curriculum was adapted to the local context and teachers, encouraged by instructions in the curriculum, gave and asked for relevant examples. Teachers in secondary schools addressed more standards, but there were fewer teachers who deliberately tried to engage learners (standard 1). Additionally, the content of some lessons seemed less relevant, since they were highly theoretical and abstract and lacked a connection or translation to students’ lives or futures.(Tables 3 and 4)

Table 3. Minimum standards addressed by primary teachers

	Motivating lessons	Mutual respect	Prior knowledge	Dialogue used	Relevant curriculum	Skills, knowledge, attitudes	Assessment	Total
Wilson				x			x	2
Guillome	x			x				2
Sylvia	x	x	x	x	x	x	x	7
Adolphe	x	x	x	x	x			5
Placedie	x		x	x	x			4
Odille	x		x	x	x	x		5
Christine	x		x	x	x	x	x	6
Azarias	x	x		x	x			4
Joseph	x	x	x	x	x	x		6
Charles	x	x	x	x	x	x	x	7
Innocent				x				1
Bernadette				x	x			2
Total	9	5	7	12	9	5	4

Table 4. Minimum standards addressed by secondary teachers

	Motivating lessons	Mutual respect	Prior knowledge	Dialogue used	Relevant curriculum	Skills, knowledge, attitude	Assessment	Total
Gabriel	x	x	x	x	x	x		6
Gaspard	x	x	x	x	x	x	x	7
Eliphas	x	x	x	x		x		5
Emile		x	x	x		x	x	5
Emmanuel		x		x		x	x	4
Christoph	x	x	x	x	x	x	x	7
Eugene	x	x	x	x	x	x	x	7
Jean Claude		x	x	x	x	x		5
Catherine			x	x	x	x		4
Jacques	x	x	x	x			x	5
Nadine			x	x	x		x	4
Paul	x	x		x			x	4
Total	7	10	10	12	7	9	8

Within groups, there was more difference among primary teachers than among secondary teachers. Some teachers demonstrated only one or two standards, whilst others addressed all. It seems plausible to assume that lessons that show many or all standards would be of higher quality, yet this was not always the case. Even though some lessons demonstrated many standards, their quality was seriously offset by lack of another standard or insufficient or incorrect subject knowledge. An example is the lesson observed in Gabriel’s classroom (male, secondary, school 5). This teacher organised a debate, preceded by a whole class question and answer session (standard 4). During this part of the lesson, the majority of students were engaged (standard 1), students contributed with their own examples (standards 2 and 3), the topic (consequences of colonialism) was relevant (standard 5) and the students could practice their communication skills (standard 6). Nonetheless, because of lack of feedback (standard 7), the content of the debate went in all directions and included some factual mistakes. Something similar happened in the lessons of Odille (female, primary, school 2) and Christine (female, primary, school 2). Classroom management issues (standard 2) hampered the quality of the lesson, despite these teachers’ various efforts to make the lesson learner-centred. These examples show that with just an inventory of the standards met, justice is not done to the complexity of teaching and the interplay of various standards. Therefore, a more in-depth description of how these teachers recontextualized LCP follows in the next part.

Constructivism

The majority of teachers applied the two central concepts of constructivism, building on prior knowledge and learning as a social interaction, in their lessons. These two concepts are embedded in standards 3 (what is taught builds on learners’ existing knowledge) and 4 (dialogue is used). To start with the first, building on prior knowledge, approximately three-quarters of the teachers asked about pre-existing knowledge or the content of the previous lesson. This should enable the learners to ‘build’ upon this knowledge and extend and deepen their current understanding (Ausubel 1968). Prior knowledge can be subdivided into academic knowledge, that is, the knowledge previously acquired in school, and what is called ‘funds of knowledge’. The latter encompasses the body of knowledge that stems from the cultural community and household that a child belongs to. These funds can become a rich source in education, provided that teachers are aware of them and are able to incorporate them in their teaching (Moll et al. 1992). One primary teacher in a rural school showcased the use of funds of knowledge when teaching about a clean environment. She brought several commonly used tools into the classroom and let her learners observe the school environment. Through visualisation and enactment and especially through reference to the learners’ own culture and context, she clarified some rather abstract concepts. This teacher unfortunately was the exception to the rule. In most primary classes, teachers perceived students’ own knowledge not only as a starting point, but also as the sole source of knowledge.

But the content of the lesson, we make sure that it is given by the students, teachers only guide the students. (Guillome, male, primary, school 1)

In secondary education, prior knowledge was applied more effectively. Most teachers started their lesson with a review of previous lessons or the homework, done either by themselves or through whole class questioning. This phase lasted in general no more than five minutes, after which the teacher moved on to the next lesson phase, which was often whole class questioning followed by a group assignment and lesson evaluation. In contrast to some primary teachers, most secondary teachers still felt that they were ultimately responsible for the acquisition of the right knowledge.

They can, they get them [skills and competences] from the preparation of the teacher …. To me, it is the teacher who first needs to be aware about the competences we want to address. If I want them to develop critical thinking, I am the one who will put them in a situation which will make them think critically. (Gaspard, male, secondary, school 5)

The second key concept of constructivism, learning as a social activity, is first and foremost manifested in group work. In all but one class, teachers put children in groups. In fact, teachers seemed to equate LCP with group work. Another approach to enhancing learning through interaction was whole class questioning, during which the teacher, with the help of hints and clues, enabled learners to discover new knowledge themselves. This approach was observed more in secondary than in primary classrooms.

At first sight, the stimulation of interaction through group work seems conducive to learning. Yet a closer look shows a less promising picture, for several reasons. First, the observations showed that the questions and assignments given to the learners nearly all targeted low cognitive levels, yielding little knowledge.

Teacher:

‘Okay, make groups of four. You have 10 minutes to discuss the questions on the chalk board’. Questions are: What is the main function of the eye? Mention two eye defects. (notes on lesson observation, Azarias, male, primary, school 3)

Additionally, one can question the efficiency of group work when learning low-complexity content. A study by Kirschner, Paas, and Kirschner (2011) confirmed that when the learning content is low in complexity, there is no additional benefit of group learning; instead, individual learning is then more effective. Third, group size, especially in primary classrooms, where class size was considerably higher than in secondary classrooms, hindered all students to participate. This was exaggerated when there were limited resources, like such as only one textbook or piece of paper to take notes on. The observations showed that when groups contained six or more students, about a third to a quarter of the pupils could not or would not participate. Lastly, for effective whole class questioning, giving appropriate feedback is key. Unfortunately, this was rarely observed in primary education. Although teachers gave ample compliments, they were often meaningless or for wrong answers. Giving compliments seemed a rather automatic act, as were the responses of the learners, who were obviously used to it. A more optimistic picture emerged from the observations in secondary education. In more than half of the lessons, the teacher either provided constructive feedback or stimulated peer feedback.

It is not good if I give them the answer directly, now let student give answer. The others correct him or her, then we conclude together, saying it is true or not. (Emile, male, secondary, school 6).

In sum, knowledge construction with the help of prior knowledge and interactions was observed in nearly all classes, yet this seemed to be done less effectively in primary than in secondary education. Teachers actively stimulated knowledge construction on the part of the learner. The findings of this study thus disagree with the assumption that LCP conflicts with traditional views on knowledge and knowledge construction whereby teachers are the sole transmitters of unquestionable knowledge (Guthrie 2018; Tabulawa 2013).

Classroom relations

According to Tabulawa (2013), one of the reasons for the failure of LCP in SSA is the social structure of African societies, which is characterised by dominance and subordinance. This description resembles Rwanda, where the leading party displays absolute control and where there is ‘an ingrained fear of authority’ (King 2014, 17). Such a social environment would conflict with the classroom relations envisaged by LCP. However, contrary to the Rwandan societal and political climate, classroom relations are best described as open, engaging and respectful (standard 2). In primary schools, some teachers deliberately tried to evoke joy and laughter. Considering the reactions of the learners, this teacher behaviour was not uncommon. Furthermore, in most classes in secondary education, and to a lesser extent in primary lessons, students posed questions and, sometimes, even dared to question comments or answers from the teacher. These new relationships were confirmed in the interviews.

Before I was like a king. I come there in front of them, I present. My job was only to deliver information. There was no friendship. But now, when I meet them outside, they are my friends. And I am their friend. Now they don’t fear a teacher. (Emmanuel, male, secondary, school 6)

The increased closeness between teacher and learner did not seem to affect teacher status. Without any exceptions, all teachers claimed that students still respected their teachers.

According to Lortie (1975), images of good education are formed by one’s own experiences as a learner, and teachers tend to copy teacher behaviours from what they experienced themselves as learners. It is therefore remarkable that the teachers in this sample did not feel hindered by their past experiences. Instead, they felt comfortable – even relieved – by the new role that is expected of them. In a recent study about pre-service teacher training in Rwanda (Iwakuni 2017), it was argued that the programme should pay more attention to student teachers’ prior perceptions as a prerequisite for effective implementation of LCP. This study shows that this shift can also be made in-service.

Language

Rwandan teachers face a double challenge regarding this educational reform: they are not only expected to master the skill of stimulating classroom interactions, they also have to do this in a language in which they are often not adequately proficient. It is therefore remarkable that only four primary and two secondary teachers mentioned language as a challenge. This might be explained by the fact that the interviews were conducted ten years after the language transfer, perhaps making English as the LoI an accepted status quo, or because teachers perceive LCP as conducive to language proficiency and students’ future.

English is important, for instance, when finding a job. They also need it when travelling to other countries. (Odille, female, primary, school 2)

Nonetheless, some teachers acknowledged that the lack of English proficiency on the part of teachers and learners has negative effects on teaching and learning. For instance, according to one primary teacher, some of her colleagues tend to fall back on traditional teaching methods such as solely providing notes. Moreover, it was reported that learners with little English knowledge do not participate: ‘If they do this in Kinyarwanda, it is very easy. But in English, the learner keeps quiet’ (Christine, female, primary, school 3). It is not hard to imagine that children who are hardly exposed to English outside school are especially the ones who tend to stay silent. Children living in urban areas and with parents who are anglophone, such as those who returned to Rwanda from neighbouring countries after the genocide, are more advantaged compared to children living in rural areas and coming from families where only Kinyarwanda (the local language) is spoken. This is confirmed by the observations: the more eloquent and assertive learners dominated group discussions. They were in charge of the limited resources and often appointed – by the teacher – as group leader.

Two coping mechanisms regarding language stood out from the lesson observations. First, half of the primary teachers regularly used code switching, that is, switching between two languages:

Teacher:

‘Now let us go back to our places. Time is over.’ Students don’t move. Teacher repeats in Kinyarwanda. Children start moving. (notes on lesson observation, Guillome, male, primary, school 1)

Code switching was not observed in secondary education, as English proficiency of students is likely to be substantially better than at the primary level. Moreover, almost half the teachers in primary and secondary education provided language support, often with the help of multisensory teaching, such as making gestures or providing illustrations.

As in some other countries, Rwandan teachers perceive English as conducive for their learners’ future and simultaneously acknowledge its negative consequences for learning and teaching. Although teachers try to counter negative effects through code switching and language support, it is likely that, as in many other African countries, mainly the already marginalised children are most negatively affected by the use of a foreign LoI (Trudell 2016).

Conclusions

This study sought to understand how effective Rwandan primary and secondary teachers recontextualize LCP. It is important to note that considering the sample size and selection criteria, the findings of this study do not allow for generalisations. Nonetheless, this study revealed some patterns and conclusions that need further consideration. First, secondary teachers seem to stimulate more knowledge construction than their colleagues in primary education. The discrepancy can be explained by the observation that secondary teachers demonstrated better questioning skills and gave more challenging assignments. This might be due to the difference in educational background; all secondary teachers in this sample had a bachelor’s degree, whilst primary teachers ‘only’ had A-level. A second explanation for the discrepancy might be their differing perception of the teacher role; primary teachers seem to have reduced their role to being a facilitator that only guides what learners themselves bring into the lesson. Secondary teachers, however, perceive it as their responsibility to provide the ‘right’ knowledge or to prompt learners to build that knowledge through stimulating interactions or challenging assignments. Hence, constructivism seems to take (deeper) root in secondary education. Further research into the reasoning and agency of teachers is needed to gain a better understanding of the observed differences.

The second conclusion concerns the assertion that LCP conflicts with the social structure of many SSA countries and thus is likely to be rejected (Tabulawa 2013; Guthrie 2018). To the contrary, evidence from this study highlighted open, joyful and respectful interactions between teachers and learners. Teachers deliberately and successfully evoked motivation and participation. Moreover, in some secondary classes, students dared to question teachers’ answers. An explanation for this remarkable finding might be the government’s attempt to encourage a spirit of individual responsibility and initiative to enhance the country’s development, in order to replace the old hierarchical system of clientelism and patronage (Honeyman 2016). This new spirit is consistent with LCP and might be strengthened by teachers’ positive perceptions of this pedagogy. In addition, according to many observers, Rwandans seems to have a general inclination to obey authorities (Uvin 1999). This obedience is also reflected in the way some teachers have responded to this educational reform; they follow governments’orders. However, there seems to be an apparent paradox between this obedience and the encouragement of one’s own initiative and critical thinking that is promoted by LCP. It is yet too early and beyond the scope of this study to examine how these new classroom relations affect and will interact with social and power structures in society, but this is undoubtedly worth future attention and research. It it also important to note once more that the sample in this study does not represent all Rwandan teachers and that perhaps, on average, teachers act in a more hierarchical manner in their interaction with learners.

The last point refers to Rwanda’s ambition to provide inclusive education for all ethnic groups. According to King (2014), in the past, education had contributed to violent intergroup conflict, since in alternating phases, access was limited or even inhibited for one of the two main ethnic groups. The fact that currently nearly all children have access to school yields promise regarding a reduction of social and ethnical divisions. Yet, as revealed by this study, LCP does not necessarily enhance learning for all. While in traditional lessons, learning for all was nearly certain not to happen, and even in well-resourced schools, learning for all is a challenging endeavour, this study highlighted two other reinforcing factors that hindered the realisation of learning for all children. First, there is the issue of language. Despite the many attempts of teachers to provide language support, some primary teachers acknowledged that insufficient proficiency on the part of teachers and learners hampered learning. With LCP this becomes more pressing, since dialogue is paramount in this approach. In addition, the majority of primary teachers interpreted LCP as being equal to group work. This interpretation, in combination with contextual factors such as lack of resources, language and large class sizes, meant a third to a quarter of the learners did not or could not participate.

In conclusion, it is worthwhile to reflect on the contribution and limitations of this study. First, similar to the work of Lattimer (2015), who also drew on Schweisfurth’s minimum standards, this study contributed to more nuanced understanding of how Rwandan teacher adapt LCP to their classroom realities. This understanding is relevant for policymakers, researches and practitioners and can support (guidance of) the implementation of LCP. However, this study revealed more. The higher number of lesson observations also showed the sometimes counter-effective interplay between the standards. Addressing nearly all of the standards but simultaneously misinterpreting or completely missing another standard can lead to lessons with little or incorrect learning. Thus, besides more examples from various contexts and education sectors, there is also a need for more research into the dynamics between the various standards. Second, this in-depth study based on interviews and lesson observations revealed differences between primary and secondary teachers, particularly regarding knowledge construction. These observed differences give an additional dimension to the debate about the feasibility of LCP in developing countries and are worth further exploration. The final reflection concerns the methodology, specifically the operationalisation of the standards. The intention behind the standards is to allow for contextual variants. However, any operationalisation that precedes the analysis includes a recontextualization. In other words, in another context the operationalisation might be different. Therefore, this study not only contributes to more insight into recontextualization itself, but also signals the need for a careful and well thought out method for analysing recontextualization.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

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