Free Primary Education and Implementation in Kenya

Free primary education policy has substantially increased school enrollment rates in Sub-Saharan African countries. The success and sustainability depend on teachers’ perception, motivation, and proper implementation of the policy in the classroom. Few studies focus on teachers’ experiences and challenges in the process of implementing the policy. The current study presents theoretical reviews using the “bottom-up,” “top-down,” and incremental policy frameworks. The study used a desk review of the documents from the Ministry of Education and other published materials from an Education Research Program in Kenya. Findings suggest that there is no adequate empirical study to support the views and assertions concerning teachers’ experiences and motivation toward the implementation of free universal primary education policy in Kenya. There is a need for empirical research and programs to understand teachers’ experiences and challenges in translating education policy into practices.


Introduction
Implementation is the nemesis of designers, it conjures up images of plans gone awry and of social carpenters and masons who fail to build to specifications and thereby distort the beautiful blue prints of progress which were handed to them. It provokes memories of "good" ideas that did not work and places the blame on second (and second-class) member of policy and administration team . . . (Honalde, 1979, p. 6) The aim of this analytical review article was to review the free primary education (FPE) policy implementation in the framework of the "bottom-up," the "top-down" (Darling-Hammond, 1990), and the interactive, incremental process (Haddad & Demsky, 1995) while placing the role of the primary school teachers in context. Top-down strategies refer to a rational, unitary, and linear process, usually from the top, whereas the bottom-up is more of a communicative process, usually from the bottom (Darling-Hammond, 1990), and the interactive model is more complex, incremental, and ongoing policy process (Haddad & Demsky, 1995). In the public policy arena, some policies require either the bottom-up, the top-down, or the incremental policy process or a combination of these approaches. This makes the policy-making process very complex, particularly in bridging the gap between policy formulation and implementation. The FPE policy initiative in Kenya was a directive implemented using the top-down model, without the application of the other two.
In this article, we review applications of the top-down approach and alternative policy frameworks in the context of FPE, in an attempt to describe the role that teachers could play in the policy process and how their participation could have impacted Kenya's FPE policy. The rest of the article is organized as follows: First, we describe the method that was used in the literature search. The "FPE" section describes the FPE policy initiative. The next three sections highlight those approaches that are often used in policy implementation, the concept of street-level bureaucracy, and the role of primary school teachers in context, respectively, followed by "Top-Down Policies and National Policy on Teacher Training." Finally, conclusions and recommendations are presented.

Method
This study was a desk review of published works from the Education Research Program (ERP) at the African Population and Health Research Center (APHRC), published literature on education in the context of Kenya, and policy documents from the Ministry of Education (MOE), Kenya. We reviewed published and working papers from ERP since 2003 when the FPE was implemented. It was important to review papers from the ERP because of the intensive research and detailed evidence on FPE, generated from demographic surveillance data. In addition, we studied the MOE policy documents on FPE, the Government of Kenya (GOK;2005a) 5-year plan (2005)(2006)(2007)(2008)(2009)(2010), and published documents from the government and by other scholars on FPE and universal primary education (UPE) outside the ERP implementation of FPE in 2003. We searched for the documents that had the following keywords: free primary education, policy, teachers, bottom up, top down, Kenya, and incremental process. We obtained the literature access to education from JSTOR, HINARI, and ERIC databases. Other literature was obtained from the APHRC publications within the Center's library. The policy documents were obtained directly from the MOE headquarters, in Nairobi.
The selection criteria were that articles were to be research articles, published reports, and government documents from the MOE since 2003 when FPE was launched in schools. These reports, research articles, and documents were those reporting on FPE as a criterion for inclusion. Synthesis was done by identifying thematic areas across the various studies and developing literature synthesis in themes. Literature synthesis is a method that has been used previously in several scientific studies (Suri & Clarke, 2009).

Free Primary Education
With the introduction of UPE in many countries in Sub-Saharan Africa (SSA), many children who had been out of school were enrolled in school and provided with the opportunity to pursue an education (GOK, 2005a;Ngware, Oketch, Ezeh, & Mudege, 2009;Ohba, 2009;. Despite the fact that the goal of many countries in SSA was to provide universal and FPE, this goal has been elusive 40 years after independence by many of these countries. With the inclusion of UPE in the Millennium Development Goals agenda, the attention toward achieving UPE has been accelerated, especially between 1990 and 2000 (Watkins et al., 2008). The accelerated effort has been demonstrated in Malawi (1994), Uganda (1997), Tanzania and Lesotho (2000), and Burundi, Rwanda, Ghana, Cameroon, and Kenya (2003) (Grogan, 2008;Kadzamira & Rose, 2003;Watkins et al., 2008).
Despite the introduction of FPE in these respective countries, research evidence shows that there has been concern with quality of instruction offered (Deininger, 2003;Oketch & Somerset, 2010), even when many governments have put a lot of emphasis on access and transition. For instance, the Kenyan government moved fast to employ more teachers to reduce class size with the numbers in employment being an outcome of class size and pupil-teacher ratio. It is expected that if FPE would be effective and sustainable in many countries, it should be a program that in the long term is accessible to all beneficiaries and provides opportunities to all schoolage children to gain access to quality education for a full cycle of basic education.
In Kenya, the 2003 FPE program was not the first initiative aimed at achieving UPE. It was first introduced in the country in 1974 when the government at the time abolished the school fees for Standards 1 to 4. The elimination of school fees was extended to Standards 5 to 7 in 1978. Subsequently, it was reintroduced in 1979 and recently in 2003. These school fee abolition initiatives had significant impact in increasing primary school enrollments, particularly for Standard 1 in 1981 (Ohba, 2009). However, scholars argue that 1 to 2 years after abolishing tuition fees in 2003, enrollments fell and dropout rates rose (Oketch & Somerset, 2010). Experts attributed this phenomenon to declining quality of education due to a massive surge in enrollment, overcrowding of classrooms, and lack of textbooks and shortage of trained teachers .

Comparisons Across Selected Countries in East and Central
Africa. In Kenya, Lesotho, Malawi, and Uganda, FPE led to a large influx of children into the schools, which resulted into an "access shock." The shock resulted into classrooms that were overcrowded, children learning in double and triple shifts, acute shortages of teachers and teaching and learning materials like textbooks (Avenstrup, Liang, & Nellemann, 2004), and large numbers of overage pupils who should have been taking adult education classes instead of being in the same class with 13-year-olds who would have been their children or grandchildren. According to Avenstrup et al. (2004), in Lesotho, Malawi, and Uganda, the public response elicited with the policy pronouncement was far greater than anticipated.
For instance, in Lesotho, enrollment into Grade 1 rose to 75% in the first year, which was far much higher than the projections by the Ministry of Education and Training, which stood at 11%. In Malawi, enrollment levels rose by 68% in 1994-the first year of FPE-yielding a gross enrollment rate of 108%. In Uganda, enrollment increased by 68% in 1 year, bringing the gross enrollment rate to 123%. However, the increase in Kenya was up by 22%, from 5.9 to 7.2 million, resulting in a gross enrollment rate of 104%. The increase in enrollment saw an increase in demand for teachers in these countries. However, there was insufficient shared knowledge in approaches for training teachers to fulfill their demand. More recently, some of these countries have adapted new methodologies, but only few educational systems in SSA countries have completely adapted teachers training to include the education for all paradigms. This is to say that teacher preparation programs encompassed the meaning and knowledge of what teachers were to expect in African classrooms in the era of UPE.

Approaches Used in Policy Formulation and Implementation
Scholars have debated the lack of clarity, non-linearity, and often indirect relationship that exists between policy guidance-as pronounced by the central authority and what happens in practice, which is often an outcome of policy implementation (Bergen & While, 2005). Therefore, the line between policy and practice is an outcome of ambiguous intentions with responses that are unpredictable (Bergen & While, 2005). The implementation of policy, which ideally should lead to effective practice, is often vague, leading to discretion being exercised by those charged with the responsibility of implementing such policies. Practitioners often take advantage of the loophole and are aided by their willingness to tailor policy guidelines to suit their practice needs. The vagueness of the policy on paper and the interpretive discretion exercised by implementers are congruent with the classical theories in policy analysis-policy implementation theory (Van Meter & Von Horn, 1975) and "street level bureaucracy" (Lipsky, 1980). Clarity of policy objectives, rules, and implementation strategies are crucial for their implementation.
Policy implementation and the factors that affect the process of implementation can be discussed using three approaches-the "top-down," the "bottom-up," and the interactive approaches commonly used in policy analysis. In top-down strategies, policy implementation is seen as a rational process, planned in advance, and strictly controlled by a central authority. Implementation requirements are outlined as a list of conditions that, if fulfilled, enables implementation to take place (Darling-Hammond, 1990;Walker & Gilson, 2004). According to rational thought in policy analysis, the gap between the policy objectives and effective implementation is seen as the result of failing to plan and control the implementation process adequately (Walker & Gilson, 2004).
Proponents of bottom-up implementation approach see policy change as an intense and communicative process. The emphasis is to understand systems of policy implementation and policy actors to understand why outcomes intended are not always achieved when policies are formulated (Darling-Hammond, 1990;Elmore, 1983;Makinde, 2005). In addition, bottom-up implementation strategies examine how the policy recipients and actors (teachers, pupils, principals, parents, and board members) view or experience the effects of the policy once it has been legislated or pronounced (Darling-Hammond, 1990). Bottom-up implementation strategies that are important for policy implementation include effective team work, focused leadership, and excellent communication. According to Darling-Hammond (1990), top-down strategies in policy implementation are not constructive; rather, they constrain the translation of policy into practice. She argues that "local agencies must adapt policy rather than adopting them . . . " (Darling-Hammond, 1990, p. 341), and when teachers and administrators have opportunities to learn constantly and make decisions during the process of implementation, the effect of the policy will persist beyond few years after the implementation. According to Rothstein (1998), bottom-up implementation strategies are regarded as important in the provision of public services like public schooling and health, which require the implementer to make discretionary decisions that in turn allow them to respond to the varied needs of their clients. For instance, a teacher in a classroom with 40 students has to find ways of balancing all the student's needs, individual ability levels, in the process of classroom instruction (Wax, 2003). Haddad and Demsky (1995) characterize education policy development and implementation as more complex, interactive, and incremental rather than centrally controlled unitary rational policy-making process. Theoretical and empirical studies on the policy-making process focus on two essential dimensions: who are involved in policy making (actors) and the process (how) of policy making (Haddad & Demsky, 1995). To capture policy complexity and its interactive nature, Haddad and Demsky (1995) combined two dimensions in a horizontal and vertical typology. The actors are placed on the horizontal axis-At one end of the spectrum is multiple actors, while on the other end is the organizational/ bureaucratic mode. The process of policy making-placed on the vertical axis-extends from the synoptic mode on the one end to the incremental approach on the other end. These two dimensions generate actor-process combinations typography. On the one extreme of this new topography is the rational model which is a composite of the synoptic method and the organizational/bureaucratic mode. On the other extreme is a composite of interaction between decision makers and implementers in a dynamic and incremental process rather than on a unitary and static policy making. Most policy making falls somewhere between these two extremes: Rigid bureaucratic approach ignores social and political context, whereas negotiated process is prone to value judgments and fluidity (Haddad & Demsky, 1995). However, in the context of Kenya, argues Sawamura and Sifuna (2008), " . . . in top down communication from the Ministry of Education to schools, government officials including district education officers did not try to listen to the voices of teachers concerned . . . " (p. 110), showing the unidimensional process in top-down policy process.
Whereas top-down and bottom-up approaches are characterized by one-dimensional process, a complex, interactive, and dynamic policy development follows a complete policy cycle. According to Haddad and Demsky (1995), this approach involves teachers-as street bureaucrats-not only being involved at the policy formulation and planning stages but also in evaluating challenges in implementation and providing feedback to improve the subsequent policy cycles. Such participation has the greatest potential to give teachers an opportunity to better understand the objectives of a policy, a sense of belonging, and be committed to the successful implementation of the policy.

The Concept of Street-Level Bureaucracy
The concept of "street-level bureaucrat" was put forward by Lipsky (1980). The main tenet of the concept is that public servants have an important role in delivering government services and "goods" to the masses. Employees at the lower level are in constant interactions with the public, and exercise discretion in execution of their work geared toward providing the needs of the masses (Bergen & While, 2005;Lipsky, 1980). Street-level bureaucrats generally perform their duties in conditions that are not favorable toward enhancing their performance. Street-level bureaucrats face high demand for their services, yet lack resources at the organizational and personal level to get the job done (Elmore, 1978;Walker & Gilson, 2004). Scholars who are proponents of the bottom-up approach argue that "the decisions of street level bureaucrats, the routines they establish, and the devices they invent to cope with uncertainty and work pressures, effectively become the public policies they carry out" (Lipsky, 1980, p. xii). According to Elmore (1978), "Street level bureaucrats are expected to treat individual as clients but the high demand for their services force them to invent routines for mass processing . . . " (p. 251). The major concern for the street-level implementer is how to control the stress and complexity of day-to-day work. According to Lipsky (1980), the day-to-day practices of street-level bureaucrats essentially become policy, not the intentions or objectives emanating from the documents and policy statements and pronouncement from the top-in many instances a centralized authority or government. For instance, what is taught to pupils in a classroom partly depends on the way a teacher delivers the curriculum in the class, which in turn depends on a teacher's level of qualification and teaching materials available to them (Ngware, Oketch, Mutisya, & Kodzi, 2010).

Placing the Primary School Teacher in Context
Teachers and classroom practices under the FPE policy. Research shows that teachers are key actors in the teaching and learning process (Oketch, Mutisya, Ngware, Ezeh, & Epari, 2010). Therefore, a policy that is focusing on the teaching and learning process should involve the teachers. However, teachers continue to be subjected to low social status, in part, because of poor pay (Bennell, 2004), and that teaching is viewed as a profession without much clout (Wiener, 2010).
Evidence shows that FPE was announced barely 1 month before the start of a school term in January 2003. Therefore, rapid implementation was the main priority, and very little time was given for consultation with teachers (Somerset, 2009). Therefore, there was little time if any for teacher induction into the new FPE policy. According to the Haddad and Demsky (1995) framework, the FPE policy went through two stages: the pronouncement of the policy decision and implementation. In so doing, key stages that would have brought more participation of the teachers were missed. These include (a) agenda setting and issue identification, (b) planning of policy implementation, (c) evaluation and modification, and (d) subsequent policy cycles.
We speculate that if the teachers were more involved in setting the agenda, planning, and evaluation, to give feedback to the system, the effect of the challenges that reduced the impact of FPE would have been minimized. Moreover, with such a declaration, what was missed was how teachers receive and translate broad policy, and curricula goals, into meaningful experiences and how these broad goals get passed to the teachers so that they can be key actors in the teaching and learning process. This reinforces our earlier assertion that the way a teacher teaches course contents are not only dependent on his/her training, but also on how he/ she internalizes the curricula, and believes in the capacity to impart knowledge to the learners. This is in turn dependent on the mechanisms with which the broad goals of the curricula content are passed down to the teacher.
This goes to emphasize that the success and sustainability of well-intended program depend on how well the objectives of the program are implemented in the classroom (Darling-Hammond, 1990). It is argued here that teachers as key players in the teaching and learning process are central in the success of Kenya's FPE. However, the disconnect between the policy-making process and the reality of implementation by teachers in schools posed major challenges in translating policy objectives into practice (Darling-Hammond, 1990;Elmore, 1983). Scholars of public policy see those charged with implementation to be at the center of the policy-making process (Darling-Hammond, 1990;Elmore, 1983), specifically teachers who are the key actors in the classroom (Darling-Hammond, 1990). Therefore, the role of primary school teachers as street-level bureaucrats cannot be overemphasized in the implementation of the FPE policy, its sustainability, and subsequent success. For instance, during the piloting of the ERP cross-sectional survey in January 2012, teachers who were participants in the focus group discussion observed that teachers are often neglected during the crucial stages of education policy process, particularly the FPE, and this can result in low teacher motivation. If teachers, who are in charge of the teaching and learning processes in the classroom, are excluded or lack necessary information and understanding, they will have little motivation to make the policy work and translate it into practice in their daily activities (Darling-Hammond, 1990;Elmore, 1983). Implementation is about capacity and motivation of implementers, the extent to which rules and regulations of implementation strategies are known by all actors, and actions taken in daily tasks done by different actors throughout the system (Darling-Hammond, 1990). We argue that the clarity with which the teachers will organize and teach math or any other subject in the schools is dependent on their experiences with the specific subjects and how well they have internalized the policy messages emanating from a policy initiative-in this case, FPE. Therefore, Wiener (2010) was right when she argued that " . . . there is little known about what challenges the policy creates, how teachers respond to these difficulties, and how this, in turn, affects students' abilities to learn" (p. 1).
Effects of FPE policy. Available evidence from the Ministry of Education, Science, and Technology (MOEST) shows that the current FPE policy has led to a significant increase in primary school enrollment in the country, from 5.9 million in 2002 to 7.2 million in 2003 (MOEST, 2004). Consequently, the number of enrolled pupils surpassed the available human and physical facilities in the 18,000 public primary schools in Kenya. Moreover, the teacher-to-pupil ratio rose from the recommended 1:40 pupils per class to 1:60 (Majanga, Nasongo, & Sylvia, 2011;Ngware, Oketch, & Ezeh, 2011). This further exacerbated the difficulty of delivering lessons in the classrooms for teachers. Large classes incapacitated the teachers' ability to organize and manage classes (Alubisia, 2005), thereby impairing the ability of teachers to provide attention to individual pupils (Wax, 2003). This led to deteriorating quality of education, one of the major challenges that eroded the initial gains and became a great concern to teachers ( Majanga et al., 2011;Ngware et al., 2011;Oketch, Mutisya, Ngware, Ezeh, & Epari, 2010;UNESCO, 2005).
At one point, teachers were asked to teach in shifts. We assert that teachers were reduced to inputs into the teaching and learning process, whose involvement in the FPE policy that was going to impact on their workloads was not necessary. Moreover, inadequate resources have compounded teachers' problems in the era of FPE, with only a half of all the classrooms in Kenya had chalkboards in the classes that were visible from all parts of the classroom (UNESCO, 2005). This, together with inadequacy of resources like learning supplies, furnishings, and appropriate infrastructure, led to teachers' incapability to fulfill their mandate in various classrooms across the country.
Large class size notwithstanding, teachers continue to grapple with increased heterogeneity among pupils in terms of age and ability. A study carried out by UNESCO (2005) showed that about 44% of the pupils who were enrolled in schools in Kenya as FPE was introduced were overage by 2 years. The foregoing discussion points to a one-sided conversation around the challenges facing teachers in the wake of FPE. These are centered on structural problems, inadequate resources, to those that stem from the presence of overage and low ability learners in the same classroom (UNESCO, 2005). However, these challenges hardly focus on teachers' roles and responsibilities as the key actors in the FPE policy implementation.
In addition, the FPE initiative was more a declaration of political expediency, with rapid implementation as the main priority (Somerset, 2009) rather than involving all education stakeholders. There was neither time to plan for implementation nor time for capacity building for teachers. The capacity building would have been in the form of in-service training for teachers to be able to cope with the complexities of increased numbers into the various schools. No policy initiative succeeds without improved capacity and motivation of those charged with the actual implementation (Darling-Hammond, 1990;Elmore, 1983). This notwithstanding, it is worth noting that FPE was a policy that elicited a lot of excitement among the population, teachers included-a policy that was meant to get all children, irrespective of their family circumstances, into school (Oketch & Somerset, 2010). The excitement was short lived for teachers.
Moreover, teachers are not getting the necessary support and guidance from local education officials to ensure that FPE succeeds. As street-level bureaucrats in the teaching and learning process, it is important to look at how teachers internalize these policies in the process of their classroom delivery and how they can adapt them to the local circumstances in the classroom setting. If teachers are not able to adapt the policy to their local circumstances, they are unable to deliver quality education to the pupils. In short, access to school does not translate into quality education if the teachers' effective control of the classroom is compromised (Abuya, Oketch, & Musyoka, 2013).
The top-down policy formulation and implementation approach to FPE did not have mechanisms to listen to the voices of teachers concerned about the declining quality of education (Sawamura & Sifuna, 2008). In general, the experiences of teachers with FPE policy in Kenya, the challenges they encounter in the classroom, their motivations and its consequence on the quality of education remains less studied. Much of what has been documented in the context of SSA relates to demand and supply of teachers as it is compromised by high costs of training (Lewin, 2002); teacher education curricula and its responsiveness to the qualities and perceptions that teachers bring into training (Coultas & Lewin, 2002); who comes for training and how teachers perceive themselves in relation to teaching, training, and future aspirations (Akyeampong & Stephens, 2002); and the fact that teachers can reflect on their experiences and produce more sophisticated accounts of learning given the right circumstances (Akyeampong, Pryor, & Ampiah, 2006).

Top-Down Policies and National Policy on Teacher Training
Scholars argue that street-level bureaucrats-in this case teachers-face high demand for their services, yet they may lack resources at the organizational and personal level to get the job done (Elmore, 1978;Walker & Gilson, 2004). This implies that teachers as key actors in policy implementation must possess the necessary pedagogical skills, be supported by the MOE at the national, local, and by their head teachers within the respective schools, to teach. Similarly, in Kenyan classrooms, the teacher must possess the pedagogical skills, the content knowledge to adequately prepare pupils in classrooms. A recent study by the APHRC indicates that teachers do not possess the necessary content knowledge to teach math curricula in the classroom (Ngware, Oketch, Mutisya, & Abuya, 2010), implying that the national teacher training curricula falls short of preparing individual teachers, to adequately teach subjects like math in the primary schools.
According to the GOK (2005b), the policy from the top as applied to primary teacher education stipulates that all primary school teachers be trained to teach all the subjects taught in the primary school curriculum in Kenya. For example, the GOK Sessional Paper No 1 states, " . . . however, the content of the entire curricula is too wide to cover while at the same time acquiring the requisite pedagogical skills . . . " (GOK, 2005b, p. 63). Thus, the training is rigorous, but the large expanse of the curricula makes it hard to cover course contents and necessary pedagogical skills required to teach the primary school curricula. A teacher trainee graduates from a primary teacher training college in Kenya ill equipped to manage large class sizes and teach students with a wide range of learning abilities. The authors argue that in such circumstances, teachers are reduced to non-actors who cannot respond to personal needs of their respective clients, nor deliver the required curricula content to the pupils for the success of FPE.
It is not surprising that the initial gains accrued to FPE in the initial 2 years have since been eroded over time (Oketch, Mutisya, Ngware, Ezeh, & Epari, 2010;Tooley, Dixon, & Stanfied, 2008) and that 60% of the pupils in the slums are attending private schools (Ngware, Oketch, Mutisya, & Abuya, 2010). Reasons that have been advanced include the perceived quality of education in the public schools and discipline that is evident among private schools (Oketch, Mutisya, Ngware, Ezeh, & Epari, 2010). The authors argue that the place of the teacher as a key actor in the classroom unable to be responsive to the needs of his learners is part of the problem of quality education in the context of FPE.
Teacher recruitment, utilization, and policy implementation. Documented evidence shows that teachers are considered educational inputs, and they are necessary for quality learning outcomes (Osei-Kofi, 2005). To this end, the Teachers Service Commission (TSC) until recently was charged with recruiting and deploying teachers. Teachers are considered as school input/non-human materials, implying that teachers' contribution to the policy is through being trained, and posted to various primary schools in Kenya. In reality, teachers can react to government policies, choose to internalize these policies in the process of their classroom teaching interactions, and adapt them to their local circumstances as frontline soldiers in the teaching and learning processes.
Therefore, as long as the teaching and learning process is looked at wholly from the point of view of "input-output," policy makers miss one of the most important ways in which policies are easily internalized-how policy recipients at the various levels (teachers, pupils, parents, school principals) experience policy pronouncements from the central government and seek to incorporate these new guidelines into their everyday work experiences. The authors argue for a change in the way teachers are viewed-not as inputs but rather as the center of the discussion about their role in the classroom, particularly during the FPE era. It is not enough to centrally pronounce policies and to centrally control teacher training and deployment, but rather it is necessary to make the teachers feel part and parcel of the policy process for which they are actors in the classroom.
Local organizational structures, practices, and policy implementation. Formerly, teachers worked under the head teacher who is the local TSC agent, answerable to the District Educational Officer, who reports to the Provincial Education Officer. Therefore, policy pronouncements pass through different levels before reaching the teacher who is the implementer in the classroom where it matters most. As such, messages that teachers get have been filtered through the different hierarchical levels. Failure of policies like FPE result from the fact that teachers are expected to construct the meanings from the broad policy for themselves, and thereby implement what can be fully understood by pupils. We hypothesize that faced with such circumstances, teachers will seek to interpret the policy in the context of their previous experiences and " . . . fill the gaps in their understanding of the policy with what is already familiar to them . . . " (Darling-Hammond, 1990, p. 342). Implicit in this is that teachers will interpret the new policy through the lens of what is familiar to them.
In addition, the voice of teachers is rarely heard by policy makers, and they are treated as passive implementers of education reforms decided at the top rather than regarding them as partners (Bennell & Akyeampong, 2007). The Kenya National Union of Teachers (KNUT) listens to the voices of teachers as concerns of their remuneration. However, little is done about the teachers' concerns and challenges that they encounter in implementing the policy in the classroom (Sawamura & Sifuna, 2008). Moreover, the short period between the announcement barely a month before the beginning of a new school term (Somerset, 2009) made the time for implementation too short for KNUT to have had adequate time to consult the teachers.
Teachers, as key actors in policy implementation, need to be empowered and possess the necessary skills and participate at the national, local, and within their respective schools' planning and decision-making processes (Wanzare & Ward, 2000). For example, teachers at the primary school level do not have continuous professional development training and workshops. In addition, primary school teachers have limited networking opportunities. This leads to lack of consultation, absence of clear guidelines, and communication strategy on implementation of policies such as FPE, which in turn has negative consequences on teachers' performance and pupils learning outcomes.
Effectiveness of the implementation of the policy is mainly focused on quantitative indicators such as financial and material inputs and increase in enrollments instead of the quality of education and learning outcomes. In addition, focusing primarily on the number of teachers and pupilteacher ratio overlooks important experiences and the voice of teachers in improving the quality of education and leaves teachers disempowered and lacking agency. To this end, Wanzare and Ward (2000) argue that involving teachers in planning, designing, and decision-making stages helps to improve teachers to articulate and understand their own training needs. In the long run, such teachers are able to internalize policy and program objectives better and ensure successful delivery and implementation (Wanzare & Ward, 2000) of policies such as FPE.

Conclusion and Recommendations
In conclusion, there is no doubt that UPE improved access of children into schools in many SSA countries, Kenya included, and enrollment rates increased (Ngware et al., 2009;Ngware, Oketch, Mutisya, & Kodzi, 2010;Ohba, 2009). We hypothesize that FPE has had significant negative professional and personal ramifications for teachers as frontline providers of the education services. Teachers were required to implement the FPE policy about which they had no clear guidelines prior to the pronouncement, let alone consulted on, and whose impact for their daily practice as teachers were largely ignored.
Moreover, providing education to a vast majority of Kenyan primary school-going children is a service. Provision of educational services depend on the internalization of the curricula and, by extension, the reaction of teachers to the policy from which the instructional materials are being drawn, how fast the teacher internalizes the contents of the policy and adapts the key components of the curriculum into tangible classroom learning experiences for the learners, and how sensitive the teacher is to the individual characteristics of the students as well as the ability of the teacher to teach a class that is organized and focused on the learning tasks. The importance of teachers as "street level bureaucrats" where the education policy matters most-in the classroom-is shared by Elmore (1983) when he posits that " . . . teachers work almost exclusively in self-contained classrooms, exercising a high degree of discretion in the management of classroom activities. Direct administrative control over classroom behavior is not only extraordinary difficult but very risky . . . " (p. 356). The assertion by Elmore (1983) sums the power of the teacher at the bottom of the pyramid in the implementation process, and emphasis on this power of the teacher is useful for the success of the provision of education service to the learners in primary schools in Kenya.
What is evident from research on the way teachers teach is that they have to be trained, materials needed for teaching are provided, and that they get regular tips on ways to organize their classrooms for implementation at the classroom level to be successful. This has to be tailored to the interaction between the teachers and their students in the classroom, and the behavior has to contribute to the teachers' sense of control in his or her classroom. In the case of FPE in Kenya, there was a sudden increase in enrollment into schools. In as much as the teachers had been trained, and the schools had some resources, local or central education officials never provided tips to teachers on the coping mechanisms with additional numbers of pupils in the wake of FPE. In addition, few questions were raised, if any, regarding the sudden announcement of a new policy, its effects on teacher-learner relationships, and how it usurped the teachers' instructional capacity and control of the classroom interaction processes (Majanga et al., 2011).
Given a history of demotivation, low self-esteem, and anger about how their profession is perceived, being ignored during the pronouncement of FPE may have added to the feeling of being unrecognized and demoralized. Moreover, FPE unanticipated and negative impact on the morale and motivation of the teachers was exacerbated by the pre-existing problems in the teaching profession, hence, undermining the their ability to provide quality education to a vast majority of Kenyan primary school children and their ability to exercise professionalism. We argue that the speed with which FPE was implemented and the failure to communicate effectively left teachers isolated in their respective classrooms as the directives and finances were channeled from the top. This ignored the aspects of the policy-making process that ought to be a more interactive and incremental rather than a centrally controlled unitary rational policy-making process (Haddad & Demsky, 1995).
Limitations of top-down policies include not allowing for identification of potential problems by the frontline providers-in this case, primary schools-on potential issues that will affect the delivery of the education service, and these policies ignore the influence of implementers' own experiences and values over the key components of policy change. Therefore, top-down policies in many instances will "constrain but not construct" practice, and local agenciesteachers in the Kenyan case-must "adapt policies rather than adopting them" (Darling-Hammond, 1990, p. 341). Teachers must be given the opportunity to learn continually, experiment with the new ideas, and make decisions during the implementation process. This, according to Darling-Hammond (1990, p. 341), determines "whether policies will come alive in schools or fade away when money or enforcement pressures end." Therefore, we acknowledge that in many instances, policy formulation is usually one single centrally controlled and completely technical decision-making process. However, for the policy cycle to be complete, the dynamic process is necessary. Consequently, there is need for composite of interaction between decision makers and implementers and interest groups in a dynamic and incremental process rather than in a unitary and static policy-making process (Haddad & Demsky, 1995).
In general, there is a widespread consensus that teachers play the key role in determining the quality of education that children receive. However, there is no adequate empirical study to support the views and assertions concerning teachers' experiences and motivation in implementing free UPE policy in Kenya. More research is needed to better understand the teachers' experiences and challenges in translating the policies into reality in the classroom teaching and learning process. Specifically, it is important to examine to what extent teachers are involved or their voices heard in policies that affect them and the learning process. It is also important to examine what areas can be strengthened in teacher training and support to cope with the dynamic nature of policies such as FPE.
This article has put forward several recommendations for moving the research forward on the role of teachers in policy implementation. First, more empirical research is needed to better understand teachers' perceptions, experiences, and challenges in translating the FPE policy into practices in their classroom teaching and learning in Kenya. Second, there is a need to develop effective communication channels not only for top-down but also bottom-up information flows between teachers and school management and other decision makers at different hierarchies of the education system. Third, engage teachers in school-level plan and decision-making process for them to have inputs in matters that affect their working conditions and to address their concerns about school management and classroom activities through dialogue. This helps to improve their sense of belonging and boost their professional commitment. Fourth, provide teachers professional support by head teachers in pedagogical activities and in managing large and heterogeneous classes. Fifth, ensure teachers professional development through in-service training and show them due respect and trust to improve their professional status and morale. It is only through greater participation of all stakeholders in the policy-making process that will serve to promote ownership of the policy and improve teachers' motivation in the FPE implementation process.

Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article

Funding
The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: This work was made possible through funding from the core support to African Population and Health Research Center (APHRC) by Rockefeller Foundation, Sida, and the William and Flora Hewlett Foundation.
Authority in Ethiopia for 6 years. His research interest includes quantitative methods, education and social stratification, child wellbeing, and adolescents' transition to adulthood.

Moses
Ngware is a research scientist and the program leader of the Education Research Program at APHRC. He holds a PhD in economics of education from Egerton University, Kenya. His current research and policy work interests are in quality and access to education in Sub-Saharan Africa, in particular, teaching and learning processes, learning outcomes, and impact evaluations of education interventions. He has published and/or contributed to more than 25 research articles in peer reviewed journals, a book chapter, and more than 20 working/discussion papers on topical issues in education.
Elijah O. Onsomu is an assistant professor of nursing at Winston-Salem State University. He received his PhD in health services research from the College of Health and Human Services at the University of North Carolina at Charlotte. He also holds a professional master in public health (MPH) from the Northwest Ohio Consortium for Public Health (NOCPH; Department of Public and Allied Health, Bowling Green State University and Department of Public Health & Preventive Medicine, College of Medicine-University of Toledo). He is also a master certified health education specialist (MCHES). He has written extensively in the areas of maternal and child health, domestic violence, HIV/ AIDS, and high-risk sexual behaviors; which have led to over 20 peer reviewed journal publications some of which on education issues.

Moses
Oketch is a reader at the Institute of Education, University of London, UK. He holds a PhD in economics of education and education policy analysis from University of Illinois at Urbana-Champaign. He previously worked at the APHRC as senior research scientist and Director of Research while on leave from the Institute of Education,University of London. He also worked as research assistant professor of education and public policy at Vanderbilt University, USA. His research interest is in economics of education with specific focus on education policy and outcomes-focusing on access, skills, livelihoods, economic development, quality, funding mechanisms and their effectiveness; policy impact assessment; policy analysis; evaluation of education programs; and general issues of educational planning.