Quality with Equity in Primary Education: Implications of High Stakes Assessments on Teacher Practice in Bangladesh

Abstract

The purpose of this study is to investigate the teacher change process and understand how the variability of primary school teachers’ in Bangladesh implementation of innovative pedagogies, such as active learning, reflects attitudes, behaviors, and concerns about the new education policy promoting high stakes exams. Qualitative data was collected from a sample of 10 teachers in 10 rural primary schools. Interviews were also conducted with staff responsible for the teachers’ professional development. Four main findings emerged. First, teachers suggested that the exams influenced their behavior and led them to question the pedagogic appropriateness of active learning. Second, the increasing focus on exam results seemed indicative of a larger trend towards placing greater value on high stakes exams. Third, exams were to serve as a common standard of outcomes for all students, yet deficiencies in technical preparation, development of concepts and methods of competency-based assessment, and clarity about purposes and use of the assessment have become obstacles to sound teacher practice and undermine the purposes of assessment. Fourth, donor support for assessment and high stakes examination at the primary level has provided legitimacy to problematic assessment policy and practices.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

A key component of quality education is ensuring positive cognitive and affective learning outcomes for students. According to Stephens (2005),

[p]olicy makers, education officials, and teachers need to have empirical evidence about students’ acquisition of knowledge, competencies, and skills, as well as factors that contribute to or detract from their acquisition, in order to incorporate this information into improved planning, management, and teaching (p. 1).

Among the earliest international policy documents advocating an increased emphasis on accountability and evaluation in low-income country contexts was the document adopted by the “World Conference on Education for All (EFA): Meeting Basic Learning Needs” in Jomtien, Thailand in 1990 (Inter-Agency Commission 1990). A decade later, at the World Education Forum in Dakar, Senegal, most governments and EFA partners reaffirmed their commitment to achieving education for all (EFA). Increasingly, across Africa and Asia, the diffusion of large-scale assessments and evaluation practices by international aid agencies is becoming a key tool for meeting this demand. Specifically, in Bangladesh there is a move within the education system to outcomes-based and standards-based education in school reform.

The requirements by government agencies for student performance coupled with mandatory testing and public reporting of results has experienced unprecedented expansion in low-income countries since the 1990s (Anderson 2008; Kamens and Benavot 2011). A critical dimension of the standards-based education movement is that schools as organizational units are held accountable for the quality of student learning, that consequences are linked to this accountability, and that schools are expected to improve where student results are below par (Anderson 2008). In 2009, the national primary education system in Bangladesh began the first stages of an attempt to assess overall levels of learning and identify differences in achievement across groups of students by introducing the primary school completion examination. Also known as the shomaponi exams (pronounced: shoma-ponee), the new public exams were the result of a major policy concern and commitment to bringing qualitative changes to the primary education system, while reducing gaps in the quality of education and increasing equity between urban and rural areas.

Across Bangladesh, primary school teachers’ concerns about the new education policy promoting high-stakes exams have been strong. This chapter focuses on the attitudes, behaviors, and concerns of a small group of primary school teachers working within a non-governmental organization (NGO) and struggling to deal with the consequences of the new public exam policy. Efforts of the well-established NGO, which runs several hundred primary schools applying the official curriculum and standards, to adopt and respond to the demands of the shomaponi exam can be regarded as typical of the problems and prospects faced by schools and teachers in Bangladesh.

This chapter is organized into four sections. In the first section, relevant literature pertaining to student assessment and evaluation in low-income country contexts is reviewed. In the second section, the national context of the study is presented including a brief overview of the land and people of Bangladesh, followed by a summary of the current state of primary education in Bangladesh. The second section ends with a description of the NGO and its primary education program, which was the research site for this study. In section three, the research design and methodology are outlined and research methodologies are justified as suitable groundings from which to collect and analyze the data. This is followed by an overview of the findings from the study with specific reference to the teachers’ ability to manage conflicting pedagogical and policy-related pressures like national assessments. Section four discusses implications of donor policy on teacher change and ongoing professional development in Bangladesh.

1 Student Assessment and Evaluation in Low-Income Country Contexts

National assessments similar to those being used in Bangladesh’s primary education system can be advantageous at different levels within the education system. At the student level, national assessments can be used to provide better details about student learning as well as to diagnose common learning problems. At the system level, large-scale assessments are a public good in that they provide transparency regarding the outputs of educational systems (Wagner 2012). This form of assessment is typically designed to evaluate the quality of the education system as a whole and curriculum delivery by schools and teachers at a particular grade level (Lockheed and Verspoor 1990). Other purposes for large-scale student assessments include support for teacher professional development and continuous improvement of instructional designs. This type of evaluation can provide useful information to education policymakers within the government and among international agencies, external donors, as well as national and international NGOs. According to Kamens and Benavot (2011), national assessments tend to be more contextually appropriate and are designed for a particular country’s intended curricular policies and educational structure.

Numerous studies point out that attempts by countries around the world to impose high-stakes examinations often lead to undesirable side effects in educational systems. National examinations place pressure on schools, administrators, teachers, and students (Keeves 1994). According to Torrance (2009) “testing does impact on the curriculum but that it narrows the curriculum to that which is tested and, in so doing, probably lowers rather than raises educational standards” (p. 84). The adverse impact on teachers entails a tendency towards preparing students for examinations and an emphasis on factual information delivered in a didactic fashion (see Goldstein 2004; Howie 2012; Tabulawa 1997). High-stakes examinations such as the shomaponi examination in Bangladesh are based on standardizing or “normalizing” students (Peters and Oliver 2009). These policies fail to recognize individual differences, talents, and achievements of students — especially those from minority groups and those with disabilities and special education needs. In addition, high-stakes assessments promote standardization of education for all students while encouraging a school climate that stigmatizes and excludes students (as well as their teachers), reinforcing an education system with the potential to increase segregation and ability tracking, as well as school drop-out rates (Peters and Oliver 2009, p. 273).

Two main reasons for assessing students are: (1) to help their learning – assessment for learning, and (2) to find out and report on what they have learned – assessment of learning. The purposes and the methods used may be different, but they are not opposed to one other. They are united by the aim of making a positive contribution to student learning. “Decisions that are involved in assessment, about the evidence to gather, how it is judged and by whom, how the results are used and by whom, follow from the reasons for the assessment” (Harlen 2007, p. 3). There is often confusion or at least lack of clarity about the reasons for assessment, and consequently, the methods used and the uses made of the assessment results. As noted above, the intended uses range from making judgements about individual pupils to monitoring accountability of the components of the education system, such as teachers and the curricular content, to evaluating effectiveness of schools or educational services in a geographical area. The use of a single assessment result for multiple purposes without adequate attention to clear definition of the objectives and matching of purposes and methods is at the root of deficiencies and abuse of student assessment in educational systems.

The term ‘high stakes’ refers to the use of student assessment results to make important decisions for the student, the teacher, the school, or for all of them (Harlen 2007). In the case of primary school students, it is indeed high stakes, when the assessment results determine whether they go to a secondary school or which one. It is worse when the child is branded as successful or a failure on the basis of the assessment, affecting a child’s self-esteem and self-confidence. The problem is compounded when the validity and reliability of the assessment is suspect.

In the more developed education systems, such as those in the United Kingdom, where public assessment of student learning has been in use and has been in development for several decades, the use of assessments and their consequences remain a continuing concern.

The consequences of pupils not achieving at certain levels can be severe, including the school being described as having ‘serious weaknesses’, being placed in ‘special measures’ or even closed. To avoid these consequences, inevitably teachers place emphasis on making sure that pupils’ test results are maximised, with all that this implies for teaching to the test and giving practice tests (Assessment Reform Group 2002, cited in Harlen 2007 p. 3).

The new primary school terminal examination in Bangladesh strives to provide a standardized and more transparent test for primary school leavers across the country. Unfortunately, a major policy concern facing educational planners and policy-makers in Bangladesh is that the intended outcomes of the shomaponi exam are not being met, which is jeopardizing efforts to improve quality and equity issues across the primary education system. The national examination results are not being used to identify weaknesses in the system such as the overwhelming use of didactic pedagogical approaches. Furthermore, the focus on examinations is narrowing teaching content and undermining teacher professionalism and the capacity for creative innovation.

A review of Bangla test question papers for the last 3 years in the primary school completion examination shows that there is little scope to assess some of the basic linguistic abilities of students such as listening, speaking, and reading skills through the test. The tests do not test at least 11 terminal competencies specified in the primary curriculum within the three broad skill areas. The analysis of test questions indicates that they do not quite match the intended terminal competencies. The tests require responses based on memorized answers from the textbook, instead of demonstration of students’ linguistic abilities. A majority of the test items are taken from exercise sections of the textbook, and learners are encouraged to memorize selected items for the test. The assessment items lead to a backwash effect on primary classroom instruction encouraging memorization and focus on low level cognitive skills rather than building the linguistic and communications foundation of students for further education and learning (Dey and Siddiqee 2012).

The national examination process appears to be having a deleterious effect on students’ educational experiences. It is commonly understood among educators in Bangladesh that the shomaponi examinations are frequently used to grade and label students and declare a large proportion of 11 and 12 year olds as “failures”. There is clearly an excessive emphasis on memorization and the creation of incentives for schools not to re-admit students who fail the examination (Swedish International Development Cooperation Agency 2011). These practices by schools have a negative effect on students’ future prospects not to mention their self-esteem. In the next section, an overview of the education policy context followed by a description of the primary education system in Bangladesh.

2 Educational Development in Bangladesh

2.1 Bangladesh: Land and People

Bangladesh, which means “Country of Bengal,” is located in the Bay of Bengal in South Asia and shares a border with India and Burma (Myanmar). The country itself is a massive delta flood plain formed by the deposit of silt from the Ganges, Brahmaputra, and Meghna rivers that pass through Bangladesh. Despite being recognized as being extremely vulnerable to climate change, Bangladesh’s population is approximately 158,570,535 making it the seventh most populous nation and one of the most densely populated counties in the world (Central Intelligence Agency 2011). Over the past two decades, amidst many odds, Bangladesh’s socioeconomic achievements have been impressive despite poor initial conditions at the time of independence and frequent exposures to severe floods and other natural disasters. Rapid economic growth and public expenditures toward social services for the poor have contributed to progress in poverty reduction. Despite such progress, Bangladesh is still one of the world’s poorest countries with 81 % of the population living below $2.00 per day (Population Reference Bureau 2011). The United Nations Development Program’s measure of poverty, the Human Development Index (HDI), has gained currency in recent years as a reliable benchmark for country comparisons based on human progress. Based on an examination of three aspects: (1) a healthy life (infant mortality, life expectancy), (2) education (adult literacy, gross enrolment ratios), and (3) standard of living (purchasing power parity, income), Bangladesh’s measure in 2011 was 0.500 placing it 146 out of 187 countries with available data (UNDP 2011).

The country has an agrarian-based economy with 76 % of the population living in rural or village settings (Population Reference Bureau 2011). Gender inequities permeate society and there is generally very low status for girls and women. Bangladesh is also characterized by linguistic, ethnic, and cultural homogeneity. The main religion practiced in Bangladesh is Islam (89.6 %), but there is a significant percentage of the population that adheres to Hinduism (9.3 %) (BANBEIS 2012). The near universality (98 %) of Bangla – the lingua franca of the country – and its use as the language of instruction in all government-run public schools is particularly advantageous for the spread of literacy and basic education (Central Intelligence Agency 2009).

2.2 Education Policy Context in Bangladesh

Modern education first developed in Bangladesh during British colonial rule in the latter half of the nineteenth century CE. This period is most notable for the widespread expansion of education among the masses. In 1854, an enquiry by the British about the state of education in the region led to the Wood’s Educational Dispatch, which recommended the promotion and provision of modern education across what is present-day Bangladesh. Some of the earliest official education policies in the form of education acts attempted to establish uniformity within the primary education system. The Primary Education Act of 1919, mandated that school management and responsibility for primary school provision was with provincial governments and the Bengal (Rural) Primary Education Act of 1930 sought to introduce universal primary education across the Bengal region (Sabur and Ahmed 2011).

During the period of Pakistani rule (1947–1971), the pace of educational development was slow and formal education remained the preserve of a select few. A 1961 population census reported that 82 % of the people of East Pakistan (present-day Bangladesh) were illiterate (Schendel 2009). Across the country, primary education suffered as the East Pakistan government was largely indifferent to providing basic education to the whole of the country. In 1971, the year of Bangladesh’s independence, evidence of the lack of headway in developing the primary education system was best illustrated by primary school participation rates reaching only 40 % (Nath and Mahbub 2008). Initiatives to increase access to primary education were generally either small-scale and local, or administratively driven from central ministries in Dhaka, with little local infrastructure in place (Unterhalter et al. 2003). In post-liberation Bangladesh, nationalization of education was the goal. The government enacted the Primary School (Taking Over) Act in 1974 that nationalized primary schools and made teachers government civil servants. Under this law, the government centralized school management thus removing the role of the district, local government bodies, and community participation in school management. According to Ahmed et al. (2007), this law failed to recognize the important role of the local community in supporting basic education:

A century’s old culture of community involvement running primary schools was effectively curbed. By implication, the law discouraged non-government providers, such as institutions run by non-government organizations (NGOs) or private providers (cited in Sabur and Ahmed 2011, p. 169).

Despite high expectations for improvements in social welfare in the post-liberation era, the first decade of independence saw limited growth or improvement in education. Although education constantly appeared on the government 5-year plans for education policy, there was little change until the mid 1980s when the government finally started to allow NGOs to provide basic primary schools in an attempt to meet the demand for places and to cover up the chronic shortfall of government provision, especially in the poorest areas.

A later attempt by the government to re-establish a modicum of decentralization into the management of primary education involved the Primary Education Act of 1981, which advocated for greater efficiency and organization in the system through provisions for the establishment of local education authorities (LEA) at the district level. This administrative reform aimed to enable the LEAs to appoint and manage teachers and supervise the functioning of primary schools, manage budgets, and conduct examinations (Sabur and Ahmed 2011). However, change of regimes hampered policy continuity and implementation. Other than a few school construction programs, primary and mass education objectives were never fully implemented in part due to President Zia’s assassination and General Ershad’s assumption of power and military rule in 1982 (Asian Development Bank 1986; Muhith 1999).

The period of greatest educational expansion coincided with the country’s first period of democratic multiparty politics. In 1990, Bangladesh adopted the Primary Education (Compulsory) Act, which entitled all children to free primary education and books. This education act was in line with the government’s commitment to the Education for All (EFA) goals adopted at the World Conference on Education for All in Jomtien, Thailand. Additional pressure on the government to improve on its primary education commitment may have come from successes of NGO interventions and increasing influence of donor support and financing in primary education provision. By the early 1990s, the largest NGO education provider alone had over one million students enrolled in over 35,000 non-formal primary schools across the country (BRAC 1996). According to Hossain et al. (2002), there were quite possibly more local NGO schools in villages scattered across rural Bangladesh than there were state schools.

The period from 1998 to 2010 has been associated with continuing government and NGO concern about access, particularly for ‘hard to reach’ children and increasingly for children that have historically been excluded from primary education including children with special needs, over-aged children, working children, street children, children from ethnic/language minority populations, and children living in remote and inaccessible areas. This period has also seen government policies that focused on issues of quality—with regard to curriculum, pedagogy, and management (Unterhalter et al. 2003). While a lack of continuity in government policy and practice has had a somewhat deleterious effect on quality education provision in Bangladesh, progress in the form of large-scale sector-wide approaches to education development have had modest successes to date.

Recently, the Government of Bangladesh formulated a New Education Policy (2010). According to the Minister of Education, “There has to be qualitative increase in both government and non-government investment and cooperation for education” (Government of Bangladesh 2010). By acknowledging the need for diversity and multiplicity in the provision of primary education, the Honorable Nurul Islam Nahid, Minister of Education, appears to provide a modicum of endorsement and recognition of NGO programs, as well as madrasas (private Islamic schools), and a variety of private and semi-private schools so long as they register and are in compliance with set rules. Other key strategies of the New Education Policy specifically addressing primary education issues include: (1) extending free and compulsory primary education from grade five to grade eight, (2) instituting a uniform curriculum and syllabus for government, non-government, private and ebtedayee ¹ madrasas (excluding quomi madrasas), (3) reducing drop-out rates among girls, children from ethnic groups, physically challenged students, and other ultra-deprived children, (4) employing interactive teaching methods, and (5) ensuring greater community participation in school development and activities. The new education policy’s emphasis on expanding access to children will likely require the government to look to alternative education providers including NGOs. The question remains to what extent the government is willing to cooperate with existing education programs beyond tacit tolerance of their existence. To achieve the goals of the latest education policy, it is recommended that the government explore alternative approaches for meaningful collaboration with non-state education providers in order to increase both access and the overall quality of primary education.

2.3 Primary Education System in Bangladesh

In Bangladesh, expansion in the variety of education providers is the result of a combination of complicated factors including: (1) the government’s inability to address the needs of population groups that are marginalized or disadvantaged for different reasons, (2) generally low-quality learning opportunities for children attending many of the government schools leading parents to look for alternatives (Rose 2002), and (3) resource and capacity constraints compelling the government to harness the support and supplementation of resources by non-state institutions (Sabur and Ahmed 2011).

The diversity of non-state primary education provision is significant. The primary education system currently comprises grades one through five, although once the new National Education Policy 2010 is implemented, the primary education program will be extended up to and including grade eight. There remain huge gaps in the government provision, however, which have led to the development within the country of different types of primary schools. Presently, the delivery of primary education occurs through a complicated system of 10 types of secular and non-secular, formal and non-formal, semi-private, private Bangla and/or English medium, and NGO-funded schools. However, the main providers are the government primary schools, registered non-government primary schools, madrasas, and NGO-operated schools (see Table 7.1). Recognizing Bangladesh is the seventh most populated country in the world, the education system must cope with extremely large numbers of students. For example, by 2010, there were over 82,000 primary schools, over 380,000 teachers, and approximately 17 million students (BANBEIS 2012). These numbers do not include thousands of private education institutions and some 40,000 non-formal primary schools operated by various NGOs, serving over 1.5 million children and poor families across the country (Ahmed 2011). These schools are generally one-room, one-teacher classes that usually offer an accelerated 4-year primary education program with learning objectives in-line with those in the national curriculum. As well, the quomi madrasa numbers are also not reported although it is estimated that the number of such institutions around the country is 15,000 and enroll more than two million students, an overwhelming majority coming from poor families in rural areas (Park and Niyozov 2008; Sikand 2004).

Table 7.1 Percentage distributions of primary school students by school type (2008)

According to the Education For All (EFA) Global Monitoring Report 2009, Bangladesh is one of the few countries in the world to have met the Dakar and Millennium Development Goal (MDG) targets of achieving gender parity in primary and secondary education by 2005—and it did so ahead of schedule (UNESCO 2010). Despite these accomplishments in education provision, Bangladesh still faces obstacles towards the long-term success of its education system.

For example:

• The rate of completion of the 5-year primary cycle varies between 50 % (Ahmed et al. 2007) and 65 % (UNESCO 2009).

• Government expenditure on education as a percentage of Gross Domestic Product (GDP) was 2.4 % in 2009, which was the lowest in South Asia (UNESCO 2011).

• At least 11 % (estimated to be 1.835 million) of children between 6 and 10 years of age are out of school and have never enrolled in any type of primary level institution (UNESCO 2009).

• The recent ratio of students to teachers in government primary schools remains high (46:1), reflecting low quality of the teaching and learning environment (UNESCO 2009).

• For the school year ending in 2004, only 56 % of primary school teachers had any formal teacher training (UNESCO 2009).

Additional statistics concerning the quantity as well as the quality of primary education in Bangladesh, as well as India, Nepal, and Pakistan can be found in Table 7.2.

Table 7.2 Primary education indicators for Bangladesh and selected countries

Even though a steady increase in enrolment rates, gender parity, and the rapid construction of new schools across the country has accelerated progress in basic education provision, formidable challenges remain for the Government of Bangladesh to overcome deficiencies in the quality of primary education.

2.4 National Assessment System

Regarding student evaluation and assessment, the New Education Policy 2010 dictates that in grades one and two there will be continuous assessments; while from grade three onwards, quarterly, half-yearly, and annual examinations will take place. Students from all types of schools across Bangladesh completing grade five will be required to take a “terminal examination” with an identical set of questions (GoB 2010). Known as the shomaponi examinations, they comprise six subjects (Bangla, English, mathematics, social studies, general science, and religion) and consist of three examinations of 2 h each, scheduled each day over a period of 3 days. Students hoping to continue their studies into secondary school (grade six) are required to pass these new high-stakes accountability examinations at the end of grade five. In an effort to address inequity in the national assessment system, since 2009 every grade five student required to write the terminal examinations also had the opportunity to earn a scholarship. Prior to the introduction of the shomaponi examination, only a select group of higher achieving students were permitted to write a separately arranged and relatively exclusive scholarship examination. Despite efforts by the government to improve the quality and increase equity within the student assessment system, the existing competency-based curriculum has some inherent limitations. For example, the New Education Policy 2010 mentions that existing evaluations and assessments of students address only the acquisition of specific knowledge and factual information. The objectives of the new examination system include: (1) a greater focus on identifying more creative evaluation methods that do not emphasize rote learning, and (2) initiatives that promote evaluations reflecting the continual growth of students’ emotional and intellectual development (GoB 2010).

3 The Study: National Assessments and Education Reform at One NGO in Northeast Bangladesh

3.1 The NGO

The study was carried out with a local NGO in the Sylhet division of northeastern Bangladesh. The NGO operates a primary education program, and at the time of the study the program consisted of 112 primary schools for children in underserved rural communities. The primary education program aims to develop and demonstrate a high-quality education model in an inclusive non-threatening classroom environment. According to one senior program officer with the NGO, “the primary education program works on a small scale but without it many children would simply not be able to get an education” (23/08/10). The NGO’s primary education program offers a pre-school program (age five) through to grade five (approximately age 10). There is a class size limit of 30 students. The primary education program conforms to the government primary curriculum from class one to class five. The schools use the government textbooks alongside supplementary materials designed by the NGO to make curriculum and lessons more child-centered and user-friendly.

The NGO aims to provide quality education in all of its schools through the use of activity-based learning materials and methods. The executive director of the NGO defines active learning as using “many varied teaching techniques, all of which follow the principle that the key to successful learning is being fully engaged and being an active partner in discovery, rather than a passive receiver of knowledge”. According to a program document on the NGO’s curriculum and methodology, “the teacher who implements active learning methodologies in the classroom must have a solid understanding of active learning principles and goals. He or she needs to be a skilled practitioner and an active partner in delivering the curriculum”.

3.2 Study Design and Research Methods

The study aimed to better understand the extent that contextual factors like national assessments exert influence on teacher concerns, mastery, and patterns of classroom innovation use. According to Datnow et al. (2002), government policies may occupy an influential place in the context of education reform, causing schools to attempt to balance multiple, and at times, conflicting demands. In the case of Bangladesh, in 2009, the Ministry for Primary and Mass Education (MoPME) implemented a new national education policy requiring all students attending government, non-government, and semi-private schools to write the shomaponi or national primary education terminal examination at the end of grade five.

For this study, qualitative data was collected from semi-structured interviews with NGO staff working for the primary education program and 10 teachers in 10 different rural primary schools operated by an NGO in northeastern Bangladesh. The focus of each interview addressed contextual conditions including the impact of assessment policies on teachers’ professional practice. The development of interview questions was generated based on a grounded theory approach to data collection and data analysis. Grounded theory can work particularly well when using semi-structured interviews as a data collection method because the flexibility of the grounded theory approach allows the researcher to develop questions and pursue areas of interest as they arise in the interview process (Kosnik et al. 2009). Both confidentiality and anonymity of participants was ensured. Participants’ real names were changed to pseudonyms and identifying geographical and or school locations were also altered.

Regarding the issue of generalizability, in qualitative studies it is normally not the goal, and therefore is not paramount (Kirby et al. 2010). This study is well positioned to discuss matters relating to a teacher’s concerns and perception of national assessments. It is important to note that the analysis and synthesis of findings is not based on data obtained from a wider population of teachers in Bangladesh, and the study is not attempting to make empirical generalizations. Nevertheless, there is no reason to assume that the sample of Bangladeshi teachers chosen and the subsequent analysis is atypical. Within the context of the NGO, teachers all come from the same region of the country, every teacher has received the same pre-service and in-service professional development and support, and the majority of teachers have grown up having experienced similar social and economic hardships.

3.3 Key Findings

For the NGO’s primary education program, the first time their students were required to sit and write the end-of-primary examinations was in 2009. Unfortunately, Bangladesh’s primary curriculum is overburdened and the end-of-cycle examinations are comprehensive, factually oriented, and not particularly child-centered. This may explain why a high proportion of the NGO’s grade five students’ terminal examination results were discouraging. The Executive Director of the NGO commented that the increasing attention and importance directed towards test scores was “a global issue” (17/08/2010). According to the Executive Director, the importance given to test scores was not an inherent objective of the NGO.

We would like to see that children are able to learn and achieve. One the one hand, students should be able to be fluent readers, able to express his or her views, able to write, able to do the basic mathematic operations well. Critical thinking, abstract reasoning skills, and being able to learn to learn. Morally, we’d like to build a kind of pluralistic, diverse, tolerant, society with different children from different ethnic communities working and living together (17/08/2010).

Having promoted a school culture of activity-based teaching and learning in which competition was carefully controlled, the Executive Director acknowledged,

[o]ur students have not done that well in this kind of competitive exam. So one of the things that we have to do is help students do better on these exams. There needs to be a balance (17/08/2010).

While acknowledging the importance of examination results, a senior leader of the NGO also described the goal of the primary education program as helping to empower and guide students in their personal and social development.

Our larger goal is to basically empower people, to empower women, men, and children. Primary education helps build that foundation of human potential. Education helps build certain human qualities and skills and an attitude and knowledge that helps in building and developing a human being to his or her full potential (17/08/2010).

Similarly, a high degree of consensus occurred among central office staff and teachers participating in the study regarding the mission or educational goals of the NGO schools. For example, while interviewing Muna who was the youngest participant in the study and also quite new to teaching with less than 2 years of experience, she stated:

[t]he main objective of this school is to educate the underprivileged children of this area, also to make them independent in the society and to prosper in future (27/07/2010).

Various teachers also mentioned that personal and social benefits were an important objective of education. Both Afzal, who has been teaching grades two and three for less than 2 years and Ritu, who has the dual role of teacher and school principal and had been with the NGO for more than 5 years, seemed to prize certain learning objectives or values related to the morality of students and the inculcation of certain life skills. For example, Afzal spoke of the importance of educating students “properly” and teaching moral values,

I mean education and being literate is not same thing. However, if they [students] are not educated properly they can’t be morally developed. Our main objective is to develop them as moral and good citizens of the country. It is our duty to present the good things to them and prevent them from the bad things (26/07/2010).

While teachers appeared to have tried to instill students with the knowledge and appreciation for being respectful and contributing members of the community, their secondary focus remained largely academic. For example, Wasifa who has been a teacher with the NGO’s education program for 7 years commented that after completing elementary school, “A student should achieve basic literacy skills as well as increased confidence” (25/07/2010). Some other experienced teachers like Hana, who has been a teacher at the NGO for 8 years, warned against encouraging students to memorize their lessons.

The government schools promote memorization. It is better to teach using a method that actively engages students. We use several materials and games to help them learn and they can easily remember the lesson (20/07/2010).

Interestingly, none of the participants in the study mentioned the importance of examination results when asked about their school’s mission or goals. One possible interpretation of these findings was that teachers did not directly associate the active learning approach with adequately preparing students to write examinations such as the shomaponi examination.

Defining and implementing a clear policy that supported a balance between outputs and a child-centered pedagogical approach appeared to be a challenge for teachers, trainers, and program officers. Parvin, a senior program officer with the primary education program, described how she was trying to decide what message she would convey to her teachers about the increasing focus on test scores.

I am looking for the answer. I think next year, maybe this time it would be more appropriate to answer. But my initial impression is that methodology is not the whole thing but the objective or goal of the organization or program is also important. Earlier we focused more on the process, not the output. If I compare with BRAC,² as I heard from people and as I know BRAC, they are more focused on output. It is the attitude of an organization, it is what they want to see. Do they want to increase the glamour of the process or are they focused on the output? I don’t care which methodology we follow but at the end of the day I want results. But I think we are implementing a very modern and very effective methodology but we need to focus on the outputs. We need to balance actually the methodology and the output (23/08/2010).

Achieving some consensus among trainers and senior program officers as to how to address the demands of the terminal examinations while simultaneously remaining faithful to the active learning approach was clearly a challenge. In particular, it seemed that the suitability of implementing an active learning methodology from preschool through to grade five was being questioned. Shafiqah, a senior teacher trainer, highlighted the challenge facing teachers and trainers.

I think that active learning should only be continued until grade three and after that we should switch to the national curriculum for students to have a shot at doing well at the grade five terminal exams (19/08/2010).

Although the new grade five terminal examinations had only been around for one academic year, it appeared that teachers were already being asked to adapt their lessons and teaching strategies, at least in grade five classes, in an effort to better prepare students. According to Azom, a senior teacher trainer with the NGO,

[t]his year, management decided ‘anyhow, anyway you will get better exam results’. In this year maybe, we are not following the teacher resource guides. This is only for grade five, next year we hope that all teacher resource guides and active learning methods and materials will be used (21/08/2010).

While interviewing the teachers sampled in the study, some suggested that the terminal examination exerted a powerful influence on their behavior and forced them to question the pedagogic appropriateness of the active learning methodology. When teachers were asked to describe in more detail the kinds of pressures they were experiencing, many commented on feeling responsible for primarily preparing students for syllabus-based national examinations. For example, Bushra, who has been teaching with the NGO for the past 7 years, spoke of her experience over the past year preparing her students for the terminal examination and pointed out her concern about test scores.

We always need to make sure that our students do better and the head office constantly checks up on us. We are always worried about how to help our students do better in the exams (19/07/2010).

The pressure felt by Bushra to increasingly focus on examination results seemed to be indicative of a larger trend among the study participants towards placing more and more value on high-stakes examinations. Muna also spoke about her desire to see her students do well on examinations and the pressure that was having on her.

At the time of students’ assessment, I want all the children to understand what I taught in my lesson. I feel pressure to teach better. At examination time, I feel pressure because I want my students to do better than students from other schools. I worry that teaching until 3:00 PM is not sufficient so I have arranged to extend classes after regular school hours (27/07/2010).

It appeared that teachers were also receiving mixed messages from their superiors. Despite some teachers being instructed during monthly meetings and other professional development workshops to focus more on adequately preparing students for their examinations, there remained an expectation that teachers would continue to faithfully follow the NGO’s mandated lesson plan guides. Parvin provided an explanation regarding the complexity of trying to accommodate the multiple demands facing the NGO’s teachers.

My senior colleagues and maybe my team members, [maybe they] are confused. I understand their confusion. Maybe they are thinking that the primary education program is going to ignore the active learning methodology in the future. Because of the shomaponi exam pressure, maybe we need to think about a more traditional approach. My idea is that the methodology was not responsible for the results. To me the challenge is to instruct them that this is a good methodology. So I am thinking about establishing more active learning methods in our program (23/08/2010).

Although Parvin, along with many of the teachers, was a strong believer in the efficacy of the NGO’s active learning pedagogy; the ongoing dialogue between teachers, supervisors, and trainers about the need to improve examination results appeared to be having an impact on teachers’ attitudes and use of active learning approaches.

During the complex process of data analysis, a number of themes or factors emerged supporting a focus on adequately preparing students for the national primary level examinations and themes or factors supporting teachers’ implementation fidelity of the NGO’s indigenous active learning model. Table 7.3 notes some of the key qualities, attitudes, and behaviors that appear to limit or support changes in teachers’ practice.

Table 7.3 Factors limiting and supporting examination preparation

4 Implications of Findings

According to Hurst and Rust (1990), efforts to improve the quality of education around the world are frequently undermined by poor working conditions for teachers including low salaries, excessive workloads, limited availability of resources, low parental involvement, and difficult environmental conditions. This is a description that fits the Bangladesh situation. The findings from this study highlight the additional challenges facing Bangladeshi educators as a result of the policy decision requiring all students completing grade five to write high-stakes national examinations.

The findings show that the study of one NGO’s struggle to deal with the consequences of high-stakes examinations was useful in exploring the policy concerns. This NGO is well known and highly regarded for its professional and technical capacity and integrity. It’s primary education program is considered by both the government and donors to be an innovative and effective model for NGO contribution to achieving national quality and equity goals in primary education. Despite the extensive experience and educational success of this NGO, it is struggling to adapt to the change introduced by education authorities in student assessment with the avowed purpose of improving quality.

The conflicting demands being placed on the teachers confirms Datnow et al. (2002) assertion that government policies force schools, and in this case the NGO and its teachers, to try and find a workable balance between policy and practice. This tension is evident in the call for change in student evaluation and examinations in the new educational policy of Bangladesh and the implementation of public examinations at the primary level.

The policy calling for changes in student evaluations and the push for more comprehensive primary school terminal examinations is embodied in the most recent National Education Policy (2010). The Minister of Education in Bangladesh promises “to ensure the rights and the opportunities for education for all…” and looks upon student evaluation as one means to achieve this goal (NEP 2010, p. 5).

In terms of international donor support for education reform in Bangladesh, the last sector-wide primary education reform called the Second Primary Education Development Program (PEDP II) was launched in 2004 and financed by a consortium of 11 donors, led by the Asian Development Bank. One key objective of PEDP II was strengthening institutional capacity within the education system including curriculum and examination reform. The Swedish International Development Cooperation Agency (SIDA), one of the international donors of PEDP II, viewed the introduction of the grade five terminal examinations across Bangladesh as a positive change. According to a SIDA (2011) the new public examination “offers a more objective means for both government, private, and non-government organization schools to prepare and assess pupils prior to moving onto secondary education. The examination also provides better protection from inconsistent or corrupt activities practiced in the past” (p. 10). The approval of high-stakes testing by the consortium of donors for PEDP II provided legitimacy to the new examination and the form it took and muted the questions and criticism about such an approach and its negative consequences noted above. The challenges facing the teachers working for the NGO highlighted in this study are indicative of the problems encountered in the education system as a whole in Bangladesh.

Among the challenges facing the NGO, it appeared that the credibility of its primary education program was becoming increasingly dependent on demonstrating that students perform well in the public examinations. The reputation of quality earned by the schools run by the NGO now depended on the majority of their students obtaining high scores in the public examinations. Yet, this new focus on examination performance appeared to involve a significant conceptual shift in the NGO’s original norms and values about quality of education. In the case of the teachers in this study, they coped with the conflicting pressures with varying degrees of success. The added challenges facing teachers were the mixed messages they were receiving from various supervisory personnel within the NGO. The apparent lack of consistency or clarity about the NGO’s expectations required teachers to try and adapt their teaching approach by trying to decide what should be continued and what is to be abandoned from familiar and well-established pedagogical practices and cultures they have acquired in the NGO.

The process by which innovations in education get adapted to varied contexts of change in which they are implemented has long been studied (Berman and McLaughlin 1976). In the case of the NGO, the process of preparing students for the shomaponi examination involved active and dynamic interactions between supervisory staff and the local educators working together within the social, organizational, and political life of the school (Datnow et al. 2002). Regardless of whether the emphasis on examination results were driven by government mandate, pressure from the international donor community, or a combination of the two, it affected the messages delivered from senior staff at the NGO about what teachers were expected to accomplish in their classrooms.

The lack of consistency about what pedagogic practices and student learning were most important meant that teachers needed to be “adaptive learning experts” (Bransford et al. 2000; Darling-Hammond 2006). Teachers had not only to use many of the child-centered teaching and learning strategies they had been trained to use but also had to exercise a relatively high level of flexibility to adapt their teaching when the existing routines did not appear to be sufficient. For example, many of the teachers interviewed commented on the difficulties they faced trying to reconcile their commitment and belief in the NGO’s child-centered instructional strategies with the growing pressure to ensure that students were adequately prepared for the national primary completion examinations. As Hattie (2009) noted, “adopting any innovation means discontinuing the use of familiar practice” (p. 252). The challenge facing teachers supports Datnow’s (2002) argument that “even when policies are seemingly straightforward, they are implemented very differently across localities, schools, and classrooms” (p. 219).

How realistic is the expectation that teachers in Bangladesh primary schools would be the “adaptive learning experts” and work out the balance between development of basic competencies through an active learning approach and the new clamor for scoring high marks in competitive public tests based on the textbook content? For a long time, the teachers in the present study appeared to measure their effectiveness in the classroom more in terms of the satisfaction and praise they received from the supervisors, trainers, principals, and parents for “doing it right” in terms of implementing active learning. As Hopkins’ (2002) observed over a decade ago, “one of the threats to child-centered learning is the narrowing of the definition of effective student learning to … test scores…” (p. 281). The school culture of the NGO has been deliberately against narrowing the definition of learning to test scores. Now they are hearing that examination scores are of paramount importance. The challenge now facing teachers, trainers, and senior education officers is to figure out alternative ways of teaching by adapting the active learning approach that they have known and used.

In terms of other more subtle factors related to the challenges facing teachers in the study is the question of teachers’ working conditions and more specifically, whether teachers are viewed and developed as “professionals”. The professional status of teachers and the drive toward greater professionalization of teachers has been the focus of extensive study (see Hurst and Rust 1990; Louis and Smith 1990; Rosenholtz 1989). Defining teachers as “professionals” in the sense of having deep curriculum and pedagogical content knowledge, a wide-ranging repertoire of instructional skills and teaching strategies, classroom management skills, the capability to accurately assess student learning and use their autonomy in the classroom to adapt their teaching and student support relative to their individual needs does not always fit with the realities of the increasingly bureaucratic and regulated conditions of most schools around the world (Louis and Smith 1990). The professional image and role of the teacher is generally regarded as too idealistic in the context of Bangladesh, but this vision has been harbored in the NGO schools at least as a longer term aspiration. The premium on test scores and the high stakes for students, teachers, and schools make it that much more difficult to nurture the professional role and image of the teacher.

Balancing an idealistic view of teaching-learning, the concept and definition of learning outcomes, and how these are measured and assessed, on the one hand, and what is pragmatically possible and achievable, on the other hand, keeping high on the agenda the principle of equity in opportunities, are challenges in all educational systems.

Among the teachers participating in this study, the majority perceived themselves to be working within a relatively restrictive structure in which they are stymied by the government’s curricular expectations and organizational constraints. The new public examination appears to have tied the teachers down further to set routines geared to test scores. So what should be and can be done? Can something positive be salvaged from the student assessment method introduced with the intention of improving the quality of student learning?

According to Spencer (2001), “the work of teachers has become deskilled through relegating teaching into a set of highly specified tasks within prescribed units of curriculum and instruction” (p. 819). This is meant to be a negative comment about common pedagogic practices. Can this structuring and creating a pattern of pedagogic behavior, in the context of numerous teaching learning constraints, including general educational qualifications and background of teachers, contribute to improved student learning and performance? The NGO’s professional development opportunities and support for teachers should be re-examined. How does this support, through a collective, coordinated, and adaptive process of change, with the aim of balancing the educational objectives of students, contribute toward achieving core competencies and performing well in test scores? The two sets of objectives can, in fact, be realized provided that the tests are designed appropriately, and used in the right way, while recognizing their limitations.

According to Hopkins (2002), an adaptive approach is more sensitive to the context of the individual school and is concerned with developing a capacity for change within the school rather than adopting a specific approach. “What is clear is that conditions need to be created within the school that ensure that individuals are supported through the inevitably difficult and challenging process of altering their ways of thinking and doing” (Hopkins 2002, p. 276). Overall, it seems that the new national examination policy for primary education had forced the NGO to reassess its existing program and consider new pedagogical reform models. It is important that the national examination model itself be critically examined, taking into consideration the generic limitations of public examinations, the specific problems of test design and construction at the primary level in Bangladesh as noted above, and the uses made of the test results for the benefit of children and improving the performance of schools, teachers, and the education system.

Some administrators appeared to support foregoing existing components of their child-centered program (at least for grade five students) while others seemed steadfast in their commitment to providing a child-centered, active learning program that supported a wide range of student learning competencies across grade levels.

5 Conclusion: Some Policy Challenges

In Bangladesh, as in many low-income countries, teachers are struggling under numerous constraints to implement new education policies advocating progressive educational reforms – one of which is the well-intentioned learning assessment approach with high-stakes public examinations at the primary level. The examinations have been introduced to serve as a common standard of outcomes for all students that could signal levels of performance of students, teachers, schools, and the system. But deficiencies in the necessary technical preparation, development of concepts, and methods of competency-based assessment at the primary level, and clarity about purposes and use of the assessment results in improving learning performance have become serious obstacles. As a result, teachers are faced with challenges that they are not equipped to tackle and can rely on little professional support and resources. Overcoming these deficiencies poses policy challenges in a number of areas described below:

• The consequences and implications of learning assessment based on summative public examinations at the primary education level have to be critically examined. The extension to the primary level of the well-recognized use of public examinations at the higher secondary level for selecting students and allocating opportunities for further education raises serious policy issues. These relate to policy and strategy concerns regarding how the tenets of universal primary education and right to education are best fulfilled. What is applicable at the secondary level are not necessarily appropriate in the same way at the primary level, both from the point of view of pedagogy for young children as well as with regard to the rights of children and obligation of state regarding basic education.

• Serious and systematic work needs to be done to make the tests valid and reliable tools for assessing basic competencies regarded as critical at the primary level. Contrary to practices in most countries where large-scale public assessments at the primary level are used, Bangladesh’s examinations are administered for all the subjects at the primary level included in the primary curriculum (and based on textbook content), rather than on core competencies, such as language and mathematical skills.

• The efficacy of any learning assessment must be judged by it consequences for students. A method that ends up declaring a large proportion of the children as failures or low performers, especially at the primary level with all the negative consequences for children’s self-esteem and motivation cannot be educationally or morally defensible. This is precisely the consequence of the Bangladeshi primary school examination in which letter grades are assigned to students and public lauding of individual students and institutions generates a highly competitive atmosphere.

• Technical and professional capacities have to be developed for continuously analyzing test domains, items, scores, and variability factors to improve validity and reliability of the tests. A major task is to move the tests away from textbook content and material and turn them into genuine measures of key competencies. Standardized expected performance levels for different grades for different competencies have to be established. How the summative assessment is linked to formative assessment in the classroom is another concern. The signals from the analysis of the test results for policy in respect of curriculum, textbooks, teaching-learning practices, teacher training, and tackling equity concerns need to be discerned through a research and analysis program. A permanent and well-resourced institutional base for technical and professional capacity building for these tasks should be established in partnership with appropriate academic institutions, rather than as a unit within the educational bureaucracy.

Examinations and tests send strong signals to the education sector, the teacher development system, and the skills required of teachers. When the examinations are of low validity, measuring rote learning rather than actual competencies, and of low reliability across time, space and socio-economic groups of examinees, and at the same time high stakes for students and the system, the overarching goal of quality with equity is liable to be seriously compromised.