Education and Urban Society - Oxfam Blogs

Educationhttp://eus.sagepub.com/ and Urban Society

Quality of Primary Education Inputs in Urban Schools: Evidence From Nairobi Moses W. Ngware, Moses Oketch and Alex C. Ezeh Education and Urban Society 2011 43: 91 originally published online 13 September 2010 DOI: 10.1177/0013124510379131 The online version of this article can be found at: http://eus.sagepub.com/content/43/1/91

Published by: http://www.sagepublications.com

Additional services and information for Education and Urban Society can be found at: Email Alerts: http://eus.sagepub.com/cgi/alerts Subscriptions: http://eus.sagepub.com/subscriptions Reprints: http://www.sagepub.com/journalsReprints.nav Permissions: http://www.sagepub.com/journalsPermissions.nav Citations: http://eus.sagepub.com/content/43/1/91.refs.html

>> Version of Record - Dec 29, 2010 OnlineFirst Version of Record - Sep 13, 2010 What is This?

Downloaded from eus.sagepub.com at UNIV CALIFORNIA BERKELEY LIB on August 2, 2012

379131

EUS

Quality of Primary Education Inputs in Urban Schools: Evidence From Nairobi

Education and Urban Society 43(1) 91–116 © The Author(s) 2011 Reprints and permission: http://www. sagepub.com/journalsPermissions.nav DOI: 10.1177/0013124510379131 http://eus.sagepub.com

Moses W. Ngware1, Moses Oketch1, and Alex C. Ezeh1

Abstract This article examines the quality of primary school inputs in urban settlements with a view to understand how it sheds light on benchmarks of education quality indicators in Kenya. Data from a school survey that involved 83 primary schools collected in 2005 were used. The data set contains information on school quality characteristics of various types of schools in Nairobi. On the basis of the national benchmarks, the quality of education provided in government schools was shown to be “better” with regard to infrastructure, teacher qualifications, and textbook provision than that provided in all the nongovernment-owned schools. However, nongovernment schools have smaller class sizes and lower pupil–teacher ratio (PTR). The bad news is that government schools have large class sizes and higher PTR and hence low levels of teacher–pupil interaction. Nongovernment schools had poor classroom structures and a higher pupil–textbook ratio, particulary private individually owned schools and community-owned schools. It also emerges that although in the government schools, student learning space is constrained by the class size, the student learning space in nongovernment

1

African Population and Health Research Center, Nairobi, Kenya

Corresponding Author: Moses W. Ngware, African Population and Health Research Center, Shelter Afrique Center, 2nd Floor, Longonot Road, Upper Hill, P.O. Box 10787, 00100—GPO, Nairobi, Kenya Email: [email protected]


92

Education and Urban Society 43(1)

schools is constrained by the classroom size. Meeting quality benchmarks in primary schooling, therefore, remains a challenge among urban populations. Keywords educational policy, school reform, urban education

Introduction The concept of “quality of education” has been difficult to define. Debate on quality of education has focused on learning achievement, relevance of the curriculum to labor markets, and/or the social, cultural, and political environment in which the learner finds himself or herself and conditions of learning including teachers and facilities (DFID, 2000; Fraumeni, Reinsdorf, Robinson, & Williams 2004; Hanushek & Luque, 2003; Hanushek & Wobmann, 2007; Lee, 2006). According to Grisay and Mahlck (1991), the notion of quality of education should go beyond student results and look at the determinants of such results including provision of teachers, buildings, equipment, and curriculum among others. From this argument, the quality of education comprises three interrelated aspects: quality of human and material resources available for teaching (inputs), quality of teaching practice (process), and the quality of results (outputs and outcomes). The first two are inputs into the schooling process. The purpose of this article is to examine the status of quality of school inputs in government-owned and nongovernment-owned schools in Nairobi settlements and to compare the observed levels of quality of inputs in the Nairobi schools with national and international standards. The article attempts to answer the following question: What is the quality of primary school inputs provided in urban schools in Kenya? The article also draws out the implications of findings in relation to this question in policy terms. Informal settlements in Nairobi are characterized by deprivation of social amenities and economic infrastructure. The housing conditions are of low quality, unplanned, and congested. Most of the people living in these informal settlements are either unemployed or do manual work or low-paying work and are classified as the low-income (poor) group. The United Nations Human Settlements Programme (UN-HABITAT) estimates that 60% of the population in Nairobi lives in informal settlements, also referred to as slums (UN-HABITAT, 2003). The formal settlements in this article mostly include locations occupied by low-middle income groups.


Ngware et al.

93

Motivation Quality education is the sixth goal of EFA, and after a period of focus on access and implementation of free primary education in Kenya, there is increasing attention being paid to the quality of “free education” or the impact of the free education on the quality of education on offer. There are several ways of assessing quality: some are based on empirical evidence, others on intuitive judgment and perceptions. What is clear according to research is that how well pupils are taught and how much they learn have an impact on the kind of interest they will have in schooling (Fehrler, Michaelowa, & Wechtler, 2007; Lloyd & Hewett, 2003; Sustainable Development Division of Human Resources and Democracy, 2001). Parents send their children to particular schools depending on judgments they make about the quality of teaching and learning provided in those schools (Guzman et al., 2008). Differences in the quality of education matter more in explaining cross-country differences in productivity growth than differences in the average number of years of schooling or enrollment rates (Barro & Lee, 1993; Glewwe & Ilias, 1996). Understanding the determination of school quality will, therefore, inform policy decisions. To create opportunities for interventions targeting school improvement, it is imperative to understand the level of quality of schooling given the rapid urbanization witnessed in developing countries. Such understanding will reveal which quality inputs require immediate intervention. Public investment in education could then target the identified school quality characteristic for intervention with a view to make the school more learner-friendly.

Conceptual Issues Quality of education can be conceptualized as the quality of the students entering the school system, quality of inputs and instructional processes, and the quality of outcomes (Bayer, Ferreira, & McMillan, 2002; Heneveld & Craig, 1995; Zhang, Postlethwaite, & Grisay, 2008). Students and school inputs interact within a social, economic, and political environment over which the school system has limited control. Many studies have used learner achievement as an indicator of educational quality (see e.g., Chinapah, 2003; Michaelowa, 2001). However, achievement results can be interpreted meaningfully only in the context of the system that produced them. To comprehensively understand and evaluate the quality of education, it is important to examine all the aspects of school quality and the ways in which environmental


94


factors both inside and outside the school community may affect quality indicators. However, this may not always be possible in a single article like this one, in addition to data limitations that sometimes limit the analysis. The rest of the article is organized as follows: Second section describes the methods and data within which the school quality indicators are assessed. Third section presents the findings and their discussion in relation to literature. Fourth, the conclusion and the implication of the findings to education policy in Kenya.

Method The Education Research Program (ERP) longitudinal study at African Population and Health Research Center (APHRC) collects data at household level in four sites in Nairobi: two informal settlements (Korogocho and Viwandani) and two formal settlements (Jericho and Harambee). To collect data on school characteristics, a school survey was carried out in 2005 under the framework of ERP. The criterion for selecting a school into the sample was primarily based on whether a school had 20 or more pupils from the sites of the longitudinal study (Korogocho, Viwadani, Jericho, & Harambee estates) enrolled and attending it. The aim was to collect information about the context, conditions, and conduct of the primary schools being attended by children from households who were participating in the ERP longitudinal survey being implemented by the APHRC. From the survey, a sample of 83 schools that had primary education as their highest level is used in the analysis. Most (70%) of the schools with primary school level had Class 8 as the highest grade, with another 7% with Class 5 as their highest grade. The highest class in the rest (13%) of the primary schools ranged from grades 1 to 7 (except grade 5). Schools in other informal and formal locations being attended by children coming from any of the four locations were also surveyed. To compare the quality of a school input with national expected standards, the article benchmarked with national norms and the International Standard Classification of Education (ISCED). National norms were based on official government reports including the Sessional paper No. 1 of 2005, the Kenya Education Sector Support Program (KESSP) document, the Staffing Norms Report, and the Handbook for inspection of educational institutions. In addition, an informal interview was held with a physical planning expert from the Ministry of Public Works who had been seconded to the Ministry of Education. The interview was mainly to confirm the norms for building facilities. The ISCED was designed by UNESCO in the early 1970s to serve “as an instrument suitable for assembling, compiling and presenting statistics of education


Ngware et al.

95

both within individual countries and internationally” (UNESCO, 1997, p. 1). On the basis of literature on effective schools and determinants of learning achievement, the Ministry of Education handbook for school inspection and other available data, we selected the inputs, described in Table 1, for analysis.

Findings and Discussion The analysis is based on 83 primary schools owned by the government (49%), community (22%), private religious group (16%), private individuals (11%), and private NGOs (2%). Of the 83 schools, 44% were in informal settlements. Apart from the Ministry of Education, schools were also registered with Ministry of Culture (28%), NGO Bureau (1%), and other unspecified institutions (1%). About 11% of the schools indicated that they were not registered. However, the religious groups that own some of the identified primary schools are small, new, and upcoming groups outside the mainstream churches, hence the use of the word “private.” There were only two private NGO-owned schools that are primarily interventions by an NGO in the provision of education in particular among marginalized communities. The word private is used to refer to the ownership of the NGO as their management seemed to be more of a social or family group, although they are expected to be charitable. Private individuals are profit-oriented investors who invest in education as a business venture to take advantage of the private demand for education. In any situation where there is a government failure in the provision of a commodity, private markets set in to fill the gap, and they are normally driven by a profit motive (Lin & Nugent, 1995).

School Size School size refers to student population within an individual school. Res earch has shown that large schools experience wider gaps in achievement between disadvantaged students and their peers (Arnold, 2000). According to Arnold, this is because they result in less communication, interaction, and coordination throughout the school, which contribute to lower student achievement. While comparing large and small schools, Arnold, Gaddy, and Dean (2004) argued that in small schools, the curriculum is more limited and directed to average pupil rather than the full range of students with varied learning needs. The distribution of schools by ownership (Table 2) shows that most of the large schools were owned by the government. This could be explained by the


96


Table 1. Indicators of Quality Education Indicator

Description of Indicator

School size Average class size PTR Pupil–toilet ratio Pupil–textbook ratio

Teacher qualifications

Quality of classroom roofing material

Quality of classroom wall material

Quality of classroom floor material

Student classroom space

Teaching load

Total number of students enrolled in a school. Average number of pupils in one classroom. This was computed by dividing the school size by number of classrooms in use for teaching. School size divided by the number of teachers in a school. School size divided by number of toilets in a school. This was further disaggregated by gender. The number of students sharing a book. This was computed for specific subjects including Science, Mathematics, and English. Average ratios are reported and disaggregated by lower (grades 1-4) and upper (grades 5-8) primary grades. Refers to the highest level of teacher training. The available qualifications for primary school teachers were untrained teacher; primary teacher 1, 2, and 3; and approved teacher, diploma teacher, and graduate teacher. The quality of material making the roof of each of classroom in a school was categorized as very good if material was tiles, good if material was iron sheets, poor if metal sheet or tin, and very poor if wood or plastic products. The quality of material making the walls in a school was categorized as very good if material was bricks or building “stones;” good if material was iron sheets; poor if cement mud or wood products; and very poor if mud, carton, or plastic. The quality of material making the floor of each classroom in a school was categorized as very good if material was cement; good if material was made of wood products; and poor if mud, dung, or soil. This was the area (measured in square meters) or physical space available to a student in a classroom. It was measured by dividing the classroom area by the number of students in the classroom. Number of hours a teacher taught in 1 week.

PTR = Pupil–teacher ratio.


97

Ngware et al. Table 2. Distribution of School Size by Ownership Statistic Observations Mean school size Standard deviation Minimum Maximum Percentage above mean

Government

Private Individual

Private Religious Private Group NGO

41 1,078

9 135

13 148

2 179

18 187

518

218

192

52

213

279 1,998 51

7 682 22

3 687 31

142 216 50

11 862 28

Community

introduction of Free Primary Education (FPE) in 2003 that attracted children to government schools. Private individuals owned the smallest-sized schools. To take advantage of economies of large scale, a Government of Kenya report recommended that in densely populated areas, schools should be expanded to 2 to 3 streams, and that the optimal primary school class size should be 45 (Teachers Service Commission, 2005). This norm was arrived at after taking into account the number of classes in a school, teaching load, regional variations, the cost implications, and the institutional arrangements. On the basis of this norm, the ideal primary school size in Nairobi should be between 720 and 1,080. In Western countries, literature indicates that an ideal school size should be between 600 and 900 students (Arnold, Gaddy & Dean, 2004). From Table 2, the mean school size was below the recommended range in all schools except in those owned by the government. However, about half of the government-owned schools had sizes above the recommended range.

Teacher Qualifications A review of more than 200 studies that examine teacher qualifications finds that teachers who are well grounded in their content areas and have greater knowledge of teaching and learning are more highly rated and more successful with students in fields ranging from basic education to vocational education (Darling-Hammond, Berry, & Thoreson, 2001; Darling-Hammond & McLaughlin, 1999). The minimum qualifications required to become a teacher in different countries can be compared using the ISCED.


98


In a number of countries, the minimum standard for primary teaching is an upper secondary qualification (ISCED 3), which is between 12 and 13 years of schooling. In Kenya, primary teacher training comes after 12 years of schooling. In few countries, the standard qualification is equivalent to a lower secondary education (ISCED 2), which is about 9 years of basic education. Such countries include Lao PDR in East Asia and the Pacific, the Maldives in South and West Asia, and a number of countries in sub-Saharan Africa (UNESCO, 2006). Teacher qualification standards at the primary level are generally lower in sub-Saharan Africa than in other regions. Except in Namibia, South Africa, and Zimbabwe, few countries require that primary teachers earn tertiary-level degrees/diplomas (UNESCO, 2006). Table 3 shows that teacher qualifications in the study schools are comparable with the ISCED standards. However, only in government-owned schools does one find that teacher qualification standards have been met. In nongovernment providers, more than 40% of teachers are not trained compared with only 1% in government-owned schools. The proportion of untrained teachers in the sample is similar (1%) to that observed in public primary schools nationally in 2005. However, government is the major employer with 75% of all teachers in the schools studied coming from government-owned schools.

Pupil–Teacher Ratio (PTR) and Average Class Size According to Benbow, Mizrachi, Oliver, and Said-Moshiro (2007), an ideal PTR should be 40:1. The debate on class size and PTR and their impact on learning achievement is not conclusive, with different studies pointing to different directions. In Ethiopia, Verwimp (1999) found a negative correlation between the quality of teaching and the PTR. However, the Ethiopian study is quick to acknowledge that class size is not a relevant variable in the quality debate whereas time allocation of teacher is. A study covering 11 of the 19 countries in the World Economic Indicators (WEI) program reports lower PTRs for the participating countries. Most WEI countries (India, Philippines, Malaysia, Sri Lanka, Tunisia, Peru, Argentina, Brazil, Chile, Paraguay, and Uruguay) had PTRs in the range of 20-30. India had the highest number (59), especially in villages (Zhang et al., 2008). Malaysia had the lowest number, with an overall PTR of 18, but 15 pupils per teacher in village schools. In Kenya, the PTR in public primary schools was 43 in 2005 and 50 in 2007 (Government of Kenya, 2008). An indication that either the number of teachers is declining or the number of pupils is growing at a much faster rate than that of teachers. The recommended PTR for public primary schools in Kenya is 40 (Teachers Service Commission, 2005). In the


99


1 8 48 40 4 100

24 8 31 1 1 65

37 12 48 2 2 100

Frequency Percentage


5 73 452 377 35 942

Private Individual

Government

31 25 46 1 0 103

30 24 45 1 0 100


Private Religious Group

8 0 5 5 1 19

42 0 26 26 5 100


Private NGO

82 19 28 3 0 132

Frequency

62 14 21 2 0 100

Percentage

Community

UT=Untrained Teacher; NACECE=National Center for Early Childhood Education; P1-P3=Primary Teacher Certificate; ATS= Approved Teacher Status; GRA=Graduate or First Degree.

UT NACECE P1–P3 ATS GRA Total

Teacher Level of Training

Table 3. Teacher Level of Training by School Ownership

100


180 160

Pupil-teacher ratio

140 120 100 80

Government

60 40

Private individual

NGO

20 Community

Private religious 0

0

10

20

30

40

50

60

70

80

90

School ownership

Chart 1. PTR by school ownership

study schools, the mean PTR ranged from 22 in schools owned by private individuals to 47 in government schools. The ratios of 20 to 30 recorded in the WEI countries compare well with what one would expect in nongovernment schools in Kenya. Schools in informal settlements had relatively high PTRs than those in formal settlement. A difference in PTR across school ownership and location is another reflection of inequality of access to quality education. Chart 1 shows the PTR across school ownership. Governmentowned schools had PTRs that were above the benchmark (see dotted line in the chart). It is notable that majority of schools in “other” formal settlement were government schools. In the study schools, the average class size ranged from 19 in schools owned by individuals to 50 in government schools. Schools in informal settlements had generally higher class sizes than those in formal settlements. Advocates of smaller class sizes argue that they provide better instruction, more individual attention to students, and accommodation to the growing diversity in schooling. Official government report recommends a class size of 45 (Teachers Service Commission, 2005). In the schools studied, more than 41% of government-owned schools had a mean class size of above 50, which was above the benchmark of 45. Chart 2 shows the distribution of average class size across school ownerships. Nongovernment schools had lower average class sizes, with most public schools having class sizes that were beyond the required benchmark. A study by Epari, Mutisya, Ezeh, Oketch, and Ngware (2008) shows that students from such private schools perform better than those from government schools in national examinations.


101

Ngware et al. 90 80

Average class size

70 Government schools

60 50 40 30

NGO

20 10 0

0

10

20

30

40

Community

Private religious

Private individual

50

60

70

80

90

School Ownership

Chart 2. Average class size by school ownership

Some studies on the effect of class size on learning achievement show positive gains on achievement in small classes. For example, the Tennesse-STAR controlled study showed that each year, the small-class students exceeded the large-class students on all cognitive and most noncognitive measures (Mosteller, 1995). According to Mosteller, gains were cumulative and stronger for students who had spent more years in small classes. The findings that class size matters have been criticized by, among others, Hanushek (1999). In his submission, Hanushek argues that small classes do not yield better student outcomes. However, scholars have dismissed his submission arguing that his analysis relied more on typical education production function studies using large and nonspecific data set not established for class-size research (Greenwald, Laine, & Hedges, 1996). In Kenya, a study by Duflo, Dupas, and Kremer (2008) shows that at the sample mean, in lower grades, reducing class size from 80 to 40 students without any other change does not lead to a significant increase in test scores. A similar finding was reported by Banerjee, Cole, Duflo, and Linden (2007) in India where no impact of the reduction in class size was achieved through the hiring of a remedial education teacher for students who remained with a regular teacher.

Safe Drinking Water and School Sanitation Facilities Safe drinking water and school sanitation facilities are critical in safe school environment as they influence children’s health and well-being. In particular,


102


girls feel unsafe to use toilet facilities that are situated in an isolated location because of the risk of rape or harassment (World Bank, 2005). In a survey carried out in South African schools in 2000, more than 30% of girls reported being raped at school (World Bank, 2005). Such a finding makes the location of toilets within a school a critical policy decision that cannot be assumed. In most schools (more than 90%), toilets were located less than 100 m from the tuition block. However, a considerable number of schools did not have separate toilets for girls and boys. For example, 47% of private individualowned schools and 32% of community schools did not have separate toilets. The Ministry of Education has set the minimum standards for the provision of toilets as part of the school sanitation facilities: the minimum number of toilets in a school is 4 for the first 30 pupils, thereafter a ratio of 25:1 and 30:1 applies for girls and boys, respectively (Government of Kenya, 2000). Going by the Ministry of Education standards, the average pupil–toilet ratios in government schools, 47 for girls and 56 for boys, were far above the benchmark. The average pupil–toilet ratio in schools owned by nongovernment entities was below 30 and within the recommended ratios. Very few schools (two government, one private religious group, and one community) had ratios that were above 100. Such schools were found in the informal settlements and other formal settlement. Charts 3 and 4 show the pupil–toilet ratios for girls and boys by school location. In general, in most of the schools visited, the average pupil–toilet ratios were above the Ministry of Education standards. In schools located in the other formal settlements, the pupil–toilet ratios were higher than in the other locations—a mean of 44 and 62 for girls and boys, respectively. Except for government schools, going by the standards, the observed average pupil–toilet ratios by school ownership and location are acceptable. Higher ratios in government schools could be explained by the relatively large school sizes. Large class sizes are as a result of excess demand brought about by the introduction of free primary education in 2003 (Oketch, Mutisya, Ngware, & Ezeh, 2008). In the study schools, water sources varied by school ownership and location. For example, the highest proportion in each of the school ownership categories by water source was 93% in government schools with piped water; 40% of private individual schools bought water from the taps, whereas another 46% had it piped to the school. Among private religious group-owned schools, 42% bought water from taps, and another 42% had piped water to school. Among community-owned schools, 37% bought water from taps, and another 63% had water piped into the school. According to the Ministry of Education, a day school should have at least 5 L of water for one child in a day and a water point for every 50 students. All the


103

Ngware et al.

160 140

Girls-toilet ratio

120 100 80 60

Viwandani Other informal Harambee

40 20 0

Korogocho 0

10

Other formal

Jericho

20

30

40

50

60

70

80

90

School location

Chart 3. Girls–toilet ratio across school locations

160 140

Boys-toilet ratio

120 100 80 60

Other informal Harambee Viwandani

40 20 0

Korogocho 0

10

20

Other fromal

Jericho 30

40

50

60

70

80

90

School location

Chart 4. Boys–toilet ratio across school locations

primary schools in the survey were day schools and, therefore, were expected to abide by this norm. The importance of quality water and sanitation facilities in schools is one of the goals of the school sanitation and hygiene education. Esrey (1994) has


104


shown that safe disposal of excreta; personal hygiene, in particular washing hands; and quantity and quality of water are most strongly related to the reduction in cases of and deaths from diarrhea among children. According to Esrey, inadequate hand washing facilities and inadequate quality of drinking water expose pupils to the threat of diarrhea. WHO (1997) argues that lack of adequate water and sanitation facilities in school creates an unsafe environment where diseases are transmitted. For example, children with worm infections have lower marks in schools than noninfected children (Nokes & Bundy, 1993; Nokes et al., 1992). This implies that children with heavy worm infections begin at a disadvantage and have a slower start in the learning process. According to Nokes et al. such children have only a few years of opportunity to benefit from a formal education as they have low school attendance.

Student Classroom Space According to UNESCO, the minimum student classroom space should be 1.5 m2 per pupil with one-seater desk, which would translate to 67.5 m2 for a room expected to hold 45 students. The Ministry of Education recommends a 7.5 m × 6.0 m classroom (Government of Kenya, 2000). This translates to 45 m2 or about 1 m2 per child in a room with 45 children. Currently, the Ministry is working on a standard of a classroom area of 61.9 m2 for 40 pupils with a one-seater desk. This will compare relatively well with the UNESCO standards of 1.5 m2 per pupil. Chart 5 shows the available student space in a classroom across schools. Most schools had student spaces that were below the required benchmark. Government-owned schools had the least student average physical space. This is largely due to the large class sizes witnessed after the introduction of FPE. Students in schools owned by private individual and private religious groups enjoyed the largest physical space in a classroom. Students in schools in the other informal settlement location had the least (mean was below 1 m2) student physical space, with schools in Korogocho having the highest (4 m2)— indicating small class sizes or large classroom space or both. The main concerns of the classroom physical space include safety and accessibility to learning, arrangement of furniture, and the teachers’ use of physical resources. Adequate space allows the teacher and students, for example, to plan a display of work, to move furniture to facilitate a group project, or to shift supplies to improve traffic flow. In a typical Kenyan classroom, desks are lined up in the traditional neat rows and facing the teacher’s desk. In a few occasions and when space allows, teachers arrange desks in


105

Ngware et al.

Student classroom space (Sq. meters)

6 5 4 3 Private individual 2 1

Private religious group private NGO

Government Community

0

0

10

20

30

40

50

60

70

80

90

School ownership

Chart 5. Student classroom space by school ownership

cooperative groups of few learners or the U-shaped configuration, where everyone has a front row seat.

Quality of Classroom Construction Materials The quality of buildings was examined by analyzing the quality of classroom roofing, walls, and floor materials of each classroom. There were 934 classrooms in the analysis—531 (57%) in government schools, 153 (16%) in private individual schools, 121 (13%) in private religious schools, 108 (12%) in community-owned schools, and 21 (2%) in private NGO schools. According to the Ministry of Education standards on school buildings, the minimum roofing material should be iron sheets gauge 26; 9-in. stones or concrete blocks for walls and a concrete slab for floor and foundation. In the schools surveyed, some of this information on buildings was difficult to collect. For example, data on the gauge of iron sheets used and the size of the building stones in schools were not captured. Roof. The quality of the roofing material was categorized as very good, good, poor, or very poor corresponding to roofing materials made of tiles, iron sheets (mabati), metal sheet/tin, and cardboards/plastic sheets. Overall, classroom roofing material was good or very good. However, for those whose roofing material was iron sheets, the quality of the sheets varied with some


106


virtual appearance looking rusty. The proportion of classrooms that had either a good or very good roof varied by ownership from 85% in private individuals to 100% in private NGO schools. It is notable that private NGO schools had the lowest (21 in total) number of classrooms, compared with 531 (in total) classrooms in government schools. Few classrooms were made of poor roofing materials—15% in private individual-owned schools, 7% in private religious groups, and 4% in government. The schools with poor or very poor classroom roofs were mainly found in Korogocho (5%) and other informal settlements (11%). Walls. The quality of the materials used in making the classroom walls was categorized as very good, good, poor, and very poor corresponding to wall materials made of bricks/stones, iron sheets (mabati), cement mud or wood/ timber, and mud/carton/plastic. Overall, 95% of classroom wall materials was good or very good. The proportion of classrooms with walls that were described as either good or very good varied by school ownership and location. For example, the walls of 97% of classrooms in government schools were described as good or very good, with only 3% being categorized as poor or very poor. In community schools, walls in 19% of the classrooms were described as poor or very poor—made up of cement-mud, wood, mud, carton, or plastic materials. In the private NGO schools, 5 of the 21 classrooms observed had walls made of cement-mud or wood. As was the case with roofing material, the schools with poor or very poor classroom walls were mainly found in Korogocho (14%) and other informal settlements (3%). Floor. The quality of the materials making the classroom floor was categorized as very good if cement; good if polished wood or wooden planks; and poor if mud, dung, or soil. Overall, classroom wall materials were very good. The proportion of classrooms with a very good floor varied by school ownership and location. For example, the condition of the floor of 96% of classrooms in government schools was described as very good, with only 2% being categorized as poor. In community schools, the floor in 91% of the classrooms was described as very good, with the remaining 9% being made of mud, dung, or soil. In the private religious group schools, 97% of classrooms had a floor with very good material with the remaining 3% being made up of mud, dung, or soil. Quite unexpectedly, 8 of the 12 classrooms observed in Harambee (a formal settlement) had floor made of poor material. Other classrooms with poor floor materials were found in Korogocho (6%) and the other formal settlements (1%). In Kenya, there are no empirical studies that have directly examined the relationship between buildings and learner achievement. However, empirical findings from the developed countries show that school’s capital


Ngware et al.

107

investment has a strong positive impact on pupil behavior and motivation with new facilities exciting the interest of pupils (Earthman & Lemasters, 1996; PriceWaterHouseCoopers, 2000). In particular, there exists close correlation between the quality of the buildings and a sense that education is important, hence increasing pupils’ interest in schooling. According to PriceWaterHouse Coopers (2000), in community schools, buildings have very marked effect on pupils’ morale. Building conditions have also been related to test scores. Literature shows that there are improvements in test scores of up to 14% when building conditions are improved (Earthman & Lemasters, 1996; Fisher, 2000).

Pupil–Textbook Ratio Textbooks are an important resource in teaching and learning. With the introduction of FPE in 2003, the Ministry of Education (MOE) provided a list of approved textbooks in all subjects. Schools following the 8-4-4 curriculum were expected to choose from the approved list of books. Approval is made based on the recommendations of the Kenya Institute of Education, which is a government agency charged with curriculum development in all public learning institutions under the MOE except Universities. The findings of this article show that in 2005, all the books used in 55% of the schools studied were approved by the MOE. The proportions of schools using all approved books ranged from 50% in community-owned schools to 71% of private religious group-owned schools. Government policy on pupil–textbook ratio stipulates that lower primary (Grades 1-4) should have a ratio of at most 3:1, whereas upper primary should have a ratio of at most 2:1 in all main subjects. The pupil–textbook ratios in Science, Mathematics, and English in both lower and upper primary grades were examined in all schools. Charts 6 and 7 show the pupil–textbook ratios by school ownership for lower and upper primary. The results show that although most government-, private religious-, and private NGO-owned schools had almost attained the required (minimum standards) pupil–textbook ratio both in lower and upper primary, other schools had relatively higher pupil–textbook ratios. For example, in the three subjects, the private individual schools had a mean ratio of 2.5 and 7.6 in lower and upper primary, respectively, whereas the community-owned schools had mean ratios of 5.4 and 2.3. Most of the schools with pupil–textbook ratios that were below the norm (which is acceptable) for lower grades were in Korogocho and other informal settlements, whereas those with higher ratios in upper grades were in informal settlements. Given that Korogocho is also an informal settlement,


108


Pupil-textbook ratio (grades 1-4)

20 18 16

Private individual

14 12 10 8 6

Government

Community

4 Private religious

2 0

0

10

20

30

40

50

60

70

80

90

School ownership

Chart 6. Pupil–textbook ratio in lower primary by school ownership

Pupil-Textbook ration (grades 5-8)

9 8 7 6 5 4

Community

3

Private NGO

2 1 0

Private Religious group Private individual

Government 0

10

20

30

40

50

60

70

80

School ownership

Chart 7. Pupil–textbook ratio in upper primary by school ownership

it seems that pupils in nongovernment schools in informal settlements had lower levels of access to Science, Mathematics, and English textbooks in school. Although in private individual schools, it is the upper grades that had ratios way above the standards; in community schools, it is the lower grades


Ngware et al.

109

that were disadvantaged. Missing out in learning experiences during the lower grades could mean that such students will be disadvantaged in terms of achievement for the rest of their schooling career.

Teaching Load Teacher workload describes the amount of time spent in teaching and interacting with pupils in and outside the classroom, the time left for preparation, and time spent in other cocurricular and management activities. The staffing norm report indicates that teachers in Kenya teach an average of 15 hr in a week compared with 23 hr in most other developing countries (Teachers Service Commission, 2005). The study measured teacher workload using the actual teaching load in hours per week. This measure has some limitations in that it did not capture the hours spent in lesson preparation. Teaching load in the schools studied varied by school ownership and location. On average, in government-owned schools, teachers taught for 32 hr in a week followed by those in private NGO (28 hr) and community and private religious groupowned schools (27 hr). Teachers who taught for more hours in a week were mostly in schools located in other formal and informal settlements. Teachers in private individually owned schools taught for an average of 17 hr in a week. From these results, teachers in government schools not only taught large class sizes but also taught for many more hours than their peers in nongovernment-owned schools. A study by Osnat, Berry, Ezeh, and Donchin (2008) shows that teachers in government schools are less motivated than teachers in nongovernment schools. Heavy workload and having to teach large classes may explain this low level of motivation.

Conclusion This article attempts to achieve two objectives: (a) to examine and present a descriptive picture of the status of quality of school inputs in urban schools from a static perspective and (b) to compare the observed levels of quality of inputs to the national and international standards. On the one hand, there is a clear evidence that government-owned schools are well endowed with textbooks in key subjects, with better buildings and more qualified teachers, and with running water and connected electricity. One of the most notable achievements of the FPE is the provision of textbooks to public schools. On the other hand, schools owned by non-state actors have a low class size that enhances teacher–pupil interactions and low PTR that gives the teacher a


110


smaller workload. The results point to an interschool inequality of educational opportunity by school ownership category. Pupils attending government-owned schools faced considerable inequality in access to child-friendly classrooms (as indicated by class size and physical space), and this becomes more pronounced in the lower grades. Pupils attending nongovernment-owned schools enjoy lower access to reading materials and quality teachers and to standard quality buildings. Given that each of these school inputs has a role to play in the quality of education, all schools owned by various providers face quality challenges. Although descriptive analysis cannot be conclusive, interschool differences in the quality of inputs originate from differences in ownership resource base and implementation of the quality and standards policy, rather than school location. For example, although 71% of the government-owned schools was visited by an education quality assurance and standards officer for at least three times in the past 12 months before data collection, only 8% of the private religious and 33% of community-owned schools had been visited—there was no such number of visits in private individual and private NGO-owned schools. It is notable that although the Ministry of Education and/or City Council of Nairobi may be aware of the physical existence of most schools, only 59% were registered with the Ministry of Education. Most of the schools not registered by the Ministry of Education miss out on the school inspection visits. It is not known how long the observed inequalities are going to exist. However, if the issues are not on the education policy agenda, then the inequalities will persist over the long term. According to Ram (1990), as the economy expands and overall access to schooling improves, educational inequality initially increases but diminishes after a certain threshold. Increased spending on public education will enhance access to schooling, but Ram’s hypothesis may not always hold as it is based on an expected macroeconomic performance whose effects in most cases take time to trickle down to the grass roots. There exists rapid urbanization in many urban centers in Africa and high levels of income inequality (a Gini index of 37 for Nairobi) in Kenya (Central Bureau of Statistics, 2005). This is reinforced by an increasing education budget (now at more than 74% of total social sector budget), most of which goes to pay salaries, and high (46%) levels of head count poverty (Central Bureau of Statistics, 2005; Government of Kenya, 2008). Such realities cast doubt on whether the Ram’s notion of a threshold can be realized. The results also show that in some cases, schools are not meeting nationally and internationally set education standards. Although teacher qualification standards are observed in government schools, the scenario was different in non-state actors-owned schools, and, in


Ngware et al.

111

particular, private individual- and community-owned schools had a considerable proportion of untrained teachers. However, the government schools operate below the nationally and internationally set standards on class size and PTR. On the availability of toilets, non-state actor-owned schools are within both the national and other international requirements, but government schools barely meet the national standards. More research needs to be done to examine the relationships between existing school quality and future aspiration of the students going through these schools. This should include the impact of the education quality on school effectiveness, teaching practice, and schooling outcomes. Kenya and many other developing countries have made strong strides toward the achievement of the goal of Universal Primary Education that strives to improve access to quality schooling for all school-going age. An important question is whether increased quantitative access can be realized without decreased quality. This article is an important initial step toward finding a solution to this critical question. From this study, several implications to education policy can be discerned. First, there is a need for the government to expand school infrastructure and to employ more teachers in urban settings to cope with rapid urbanization and to create opportunities for disadvantaged urban children to access quality education. This includes the aim to reduce both class sizes and PTRs to the Ministry of Education’s acceptable standards. This may call for radical measures some of which may require political will. Schools require extra grounds for expansion, and this puts them into conflict with neighboring landlords among other interest groups. In public interest, it is plausible to include school expansion projects in the ongoing slum upgrading program. However, there should be wider consultation with neighboring communities to explain both the short-term and long-term benefits of the school expansion. Second, to improve the quality of education in private individually owned and in community-owned schools, public–private partnership initiatives should be encouraged and supported by the government. For this to be successful across different school ownerships, all education institutions should register with the Ministry of Education as they stand to benefit from technical support among other gains. Other benefits include the provision of textbooks by the government to pupils attending privately owned schools but belonging to underprivileged populations if such a recommendation is implemented. Inspection visits need to be scaled up in non-state actors-owned schools. The need to extend the supply of school inputs to private schools is strengthened by the fact that children are expected to be treated equally (see the Children Act 2001). However, this will require effective targeting mechanisms and the ceding of more control to the government in the running of private and


112


community schools. Finally, schools exist within communities with stakeholders outside the education sector as key players in the development in informal settlements. In view of this, it is imperative to address the problem of school inputs from a sectorwide approach. For example, provision of school toilet facilities and safe drinking water would benefit from a sectorwide approach with stakeholders coming from the education, housing, environment, health, water, and local government subsectors. Author’s Note The content and views in this paper are the sole responsibility of the authors and do not necessarily reflect the views of any persons mentioned or the funding organization.

Acknowledgment We acknowledge the important contribution of APHRC staff who participated at various stages of the development of this article including data collection and processing and giving valuable comments during the internal review process. We are also grateful to our partners including the City Council of Nairobi Department of Education for providing useful information.

Declaration of Conflicting Interests The authors declared that they had no conflicts of interests with respect to their authorship or the publication of this article.

Funding Funding for this study was provided by The William and Flora Hewlett Foundation through the Education Research Program at APHRC.

References Arnold, M. L. (2000). Rural schools: Diverse needs call for flexible policies [policy brief]. Aurora, CO: Mid-continent Research for Education and Learning. Arnold, M. L., Gaddy, B. B., & Dean, C. B. (2004). A look at the condition of rural education research: Setting a direction for future research. Aurora, CO: Midcontinent Research for Education and Learning. Banerjee, A., Cole, S., Duflo, E., & Linden, L. (2007). Remedying education: Evidence from two randomized experiments in India. Quarterly Journal of Economics, 122, 1235-1264. Barro, R. J., & Lee, J. (1993). International comparisons of educational attainment. Journal of Monetary Economics, 32, 363-394.


Ngware et al.

113

Bayer, P., Ferreira, F., & McMillan, R. (2002). A unified framework for measuring preferences for school quality. Retrieved October 17, 2008, from http://www .econ.yale.edu/seminars/labor/lap02/bayer-021018.pdf Benbow, J. Mizrachi, A., Oliver, D., & Said-Moshiro, L. (2007). Large class sizes in the developing world: What do we know and what can we do? American Institutes for Research under the EQUIP1 LWA, U.S. Agency for International Development. Central Bureau of Statistics. (2005). Geographic dimensions of well-being in Kenya: Who and where are the poor, a constituency level profile (2nd Vol.). Nairobi: Central Bureau of Statistics. Chinapah, V (2003). Monitoring learning achievement (MLA) project in Africa. Association for the Development of Education in Africa ADEA Biennial Meeting 2003 (Grand Baie, Mauritius, December 3-6, 2003). Association for the Development of Education in Africa (ADEA), International Institute for Educational Planning. Darling-Hammond, L., Berry, B., & Thoreson, A. (2001). Does teacher certification matter? Evaluating the evidence. Educational Evaluation and Policy Analysis, 23, 57-77. Darling-Hammond, L., & McLaughlin, M. W. (1999). Investing in teaching as a learning profession: Policy problems and prospects. In L. Darling-Hammond & G. Sykes (Eds.), Teaching as the leaning profession. A handbook of policy and practice (pp. 376-411). San Francisco, CA: Jossey-Bass. DFID. (2000). The quality of learning and teaching in developing countries: Assessing literacy and numeracy in Malawi and Sri Lanka. (DFID, Education Research Paper No. 41). Duflo, E., Dupas, P., & Kremer, M. (2008). Peer effects, class size, and teacher incentives: Evidence from a randomized evaluation in Kenya. Available from http:// vle.worldbank.org/bnpp/files/TF056474KenyaETP061607.pdf Earthman, G., & Lemasters, L. (1996). Review of research on the relationship between school buildings, student achievement, and student behaviour. A paper presented at the Annual Meeting of the Council of Educational Facility Planners. Tarpon Springs, Florida, USA, October 8, 1996. Epari, C., Mutisya, M., Ezeh, A., Oketch, M., & Ngware, M. (2008). Factors associated with low achievement among students from Nairobi’s urban informal neighbourhoods. (African Population and Health Research Centre, Working Paper No. 41). Esrey, S. (1994). Complementary strategies for decreasing diarrhoea morbidity: Water and sanitation. Paper presented at the Pan American Health Organisation, March 2-3, 1994. PAHO Head Office, Washington, D. C., USA. Fehrler, S., Michaelowa, K., & Wechtler, A. (2007, May). The cost-effectiveness of inputs in primary education: Insights from recent student surveys for sub-Saharan Africa. Proceedings of the German Development Economics Conference, GÃttingen, Verein für Socialpolitik: Research Committee Development Economics.


114


Fisher, K. (2000). Building better outcomes. The impact of school infrastructure on student outcomes and behaviour, schools issues digest. Department of Education, Training and Youth Affairs, Australian Government. Available from http://www.dest.gov .au/NR/rdonlyres/69728B50-143A-4C9A-BD91-3703FC7A80DC/4507/building.pdf Fraumeni, B. M., Reinsdorf, M. B., Robinson, B. B., & Williams, M. P. (2004, June). Price and real output measures for the education function of government: Exploratory estimates for primary & secondary education. Paper presented at the Conference on Research in Income and Wealth Conference on Price Index Concepts and Measurement, Vancouver, BC, Canada. Glewwe, P., & Ilias, N. (1996). The determinants of school Africa: A case study of Ghana attainment in sub-Saharan, Journal of International Development, 8, 395-413. Government of Kenya. (2000). Handbook for inspection of educational institutions. Nairobi: Ministry of Education, Science and Technology. Government of Kenya. (2008). Education sector report 2008: Realizing vision 2030 goals through effective and efficient public spending. Nairobi: Ministry of Education. Greenwald, R., Laine, R. D., & Hedges, L. W. (1996). The school funding controversy: Reality bites. Education Leadership,53, 78-79. Grisay, A., & Mahlck, L. (1991). The quality of education in developing countries: A review of some research studies and policy documents. Paris: International Institute of Educational Planning. Guzman, A. B., Castro, B. V., Aquino, K. A. B., Buenaventura, M. A. R., Duque, A.C.C., & Enriquez, M. L. D. R. (2008). Filipino parents’ school choice and loyalty: A factor analysis. Educational Research for Policy and Practice, 7. Retrieved October19, 2008, from Education Research for Policy and Practice, 7, 109-122. Hanushek, E. A. (1999). Some findings from an independent investigation of the Tennessee STAR experiment and from other investigation of class size effects. Educational Evaluation and Policy Analysis, 21, 143-163. Hanushek, E. A., & Luque, J. A. (2003). Efficiency and equity in schools around the world, Economics of Education Review, 22, 481 -502, Available from http:// edpro.stanford.edu/hanushek/admin/pages/files/uploads/hanushek+luque .eer.2003.pdf Hanushek, E. A., & Wobmann, L. (2007). Education quality and economic growth. Washington, DC: The International Bank for Reconstruction and Development/ The World Bank. Heneveld, W., & Craig, H. (1995). Schools count: World bank project designs and the quality of primary education in Sub-Saharan Africa. (World Bank Technical Paper No. 303). Lee, M. P. (2006). Widening gap of educational opportunity? A longitudinal study of educational inequality in China. (UNU-WIDER, Research Paper No. 2006/66).


Ngware et al.

115

Lin, J. Y., & Nugent, J. B. (1995). Institutions and economic development. In J. Behrman & T. N. Srinivasan (Eds.), The handbook of development economics (Vol IIIA), 2301-2370). Amsterdam: Elsevier. Lloyd, C. B., & Hewett, P. C. (2003). Primary schooling in sub-Saharan Africa: Recent trends and current challenges( WP No. 176). Social Science Research Division, Population Council. Retrieved October15, 2008, from http://www .popcouncil.org/pdfs/wp/176.pdf Michaelowa, K. (2001). Primary education quality in Francophone Sub-Saharan Africa: Determinants of learning achievement and efficiency considerations. World Development, 29, 1699-1716. Mosteller, F. (1995). The Tennessee study of class size in the early school grades. The Future of Children: Critical Issues for Children and Youths,5, 113-127. Nokes, C., & Bundy D. (1993). Compliance and absenteeism in schoolchildren: Implications for Helminth control. Transactions of the Royal Society of Tropical Medicine and Hygiene, 87, 148-1521. Nokes, C., Grantham-McGregor, S. M., Sawyer, A. W., Cooper, E. S., Robinson, B. A., & Bundy, D. A. (1992). Moderate to heavy infections of trichuris trichiura affect cognitive function in Jamaican school-children. Parasitology, 104, 539-547. Oketch, M., Mutisya, M., Ngware, M., & Ezeh, A. C. (2008). Why are there proportionately more poor pupils enrolled non-State Schools in Urban Kenya in spite of FPE Policy? (African Population and Health Research Centre, Working Paper No. 40). Osnat, K., Berry, E. M., Ezeh, A., & Donchin, M. (2008). Determining appropriate entry point for health promoting schools intervention in Nairobi informal settlements. (African Population and Health Research Centre, Working Paper No. 42) PriceWaterHouseCoopers. (2000). Building performance: An empirical assessment of the relationship between schools capital investment and pupil performance. United Kingdom: Department for Education and Skills. Ram, R. (1990). Educational expansion and schooling inequality: International evidence and some implications’. Review of Economics and Statistics, 72, 266-274. Sustainable Development Division of Human Resources and Democracy. (2001). Basic education in sub-Saharan Africa issue briefs from USAID’s Africa Bureau Office of Sustainable Development Division of Human Resources and Democracy. Retrieved October 1, 2008, from http://www.equip123.net/EQ_ Review/docs/BasicEdBriefs.pdf Teachers Service Commission. (2005). Study on teacher staffing norms (3rd Draft Report). Nairobi: Ministry of Education science and Technology. UNESCO. (1997). International standard classification of education ISCED 1997. Retrieved September 30, 2008, from http://www.unesco.org/education/information/ nfsunesco/doc/isced_1997.htm


116


UNESCO. (2006). Teacher training, qualifications and education quality. Retrieved October 8, 2008, from http://www.uis.unesco.org/TEMPLATE/pdf/Teachers2006/ Teachers_Ch2.pdf UN-HABITAT. (2003). The challenge of slums, Global report on Human Settlements 2003. Nairobi. Available from http://www.unhabitat.org/downloads/docs/ GRHS.2003.0.pdf Verwimp, P. (1999). Measuring the quality of education at two levels: A case study of primary schools in rural Ethiopia. International Review of Education, 45, 167-196. WHO. (1997). Primary school physical environment and health: WHO global school health initiative. Geneva: WHO Information Series on School Health: Document Two. World Bank. (2005). Hygiene, sanitation and water in schools: Gender roles and impact. Retrieved September 24, 2008, from http://www.schoolsanitation.org/ BasicPrinciples/GenderRoles.html Zhang, Y., Postlethwaite, T. N., & Grisay, A. (eds). (2008). A view inside primary schools: A world education indicators (WEI) cross-national study. Montreal: UNESCO Institute for Statistics.

Bios Moses W. Ngware is an associate research scientist at the African Population and Health Research Center (APHRC). He is a project manager in the Education Research Program and oversees the design and implementation of education research projects. His current research focuses on teaching practice and opportunities to learn. He holds a PhD in economics of education from Egerton University (Kenya). Moses Oketch is a senior research scientist at APHRC and is on sabbatical leave from The Institute of Education, University of London. He is the theme leader of the Education Research Program at APHRC. His research interest is in economics of education with specific focus on education policy and outcomes. He holds a PhD in economics of education and education policy analysis from University of Illinois at Urbana-Champaign (United States). Alex C. Ezeh is the executive director of the African Population and Health Research Center and honorary professor of public health at the University of the Witwatersrand, South Africa. His research interest focuses on demographic patterns, health outcomes, and their linkages in Africa. He leads a new consortium to strengthen doctoral training in Africa and serves on the boards and committees of several international public health organizations. He holds a PhD in demography from the University of Pennsylvania (United States).
