Communities, gender and education: Evidence from sub-Saharan Africa
Background paper for 2003 UNESCO Global Monitoring Report
Pauline Rose
Centre for International Education, University of Sussex
p.m.rose@sussex.ac.uk
Introduction
Community participation has received increased attention in international and national policy
in recent years. It is considered important as an end in itself (as a democratic right), as well as
a means to the achievement of sustainable development and poverty alleviation (Stiglitz,
1997). The interest in community participation has occurred simultaneously with an
intensified focus on achieving gender parity in education, and community participation may
be seen as one of the means to achieve this goal. One of the potential outcomes of community
participation as an end in itself is the transformation of gender relations, allowing the
opportunity for women to participate alongside men in decision-making, for example. As a
means to an end, community participation in education is seen as a way to increase resources,
improve accountability of schools to the community they serve, ensure a more cost-effective
use of resources and, importantly, be responsive to local needs. As a result, it intends to
improve equitable access, retention, quality and performance of schooling.
Table 1: Matrix of dimensions and degrees of participation in education
Forms of
Pseudoparticipation participation
Education
functions
Use of
service
Genuine
participation
Contribution Attendance at Consultation Involvement
of resources
meetings
on issues
in delivery
Delegated ‘Real’ powers
powers and and decisiondecisionmaking
making
Designing
policy
Curriculum
development
Teacher
hiring/firing
Supervision
Payment of
teachers
Teacher
training
Textbook
design
Textbook
distribution
Certification
Building and
maintenance
Mobilising
resources
Source: Adapted from Bray 2000 (Table 1, p. 20).
1
There are potentially a range of areas in which communities can be involved in education,
from mobilisation of resources and constructing classrooms, to supporting the development of
curriculum and design of policy. In addition, different degrees of participation might be
apparent, from ‘pseudo-participation’ of community members in terms of their use of a
service and contribution of resources, to their ‘genuine-participation’ in decision-making
(Table 1). The degree of participation can vary between communities as well as by different
members within communities. As explored below, women and men may participate in
different ways and in different types of activities.
This paper begins by exploring the different ways in which communities are encouraged to
participate in education. This is followed by a review of the extent to which community
participation has contributed to improving gender equity in educational outcomes in subSaharan Africa (SSA). Finally, it examines whether community participation in education, as
an end in itself, is contributing to the transformation of gender relations.
Forms of community participation in education
While community participation in schooling has always been apparent in many countries in
SSA, particularly in the form of support for school construction, it has become more
formalised in policy in recent years with new forms of community participation emerging.
The paper will consider four main types of community participation which are evident in
education – spontaneous community schools, international agency-supported community
schools (integrated and alternative forms), community participation in government schools,
and community participation in cross-sectoral programmes with implications for education.
Spontaneous community schools
Community schools involve the community in construction and management of schools,
although the extent to which the community is involved can vary considerably. The
establishment and support of schools by communities has always been evident in many SSA
countries, often as a response to the failure of government provision. In Kenya, for example,
the secondary system evolved largely as a result of community support through Harambee
schools. These are seen as one of closest examples in SSA to ‘spontaneous grassroots
initiative for the delivery of education’ (Rugh and Bossert 1998: 36). However, over time, the
lower quality of these schools compared with government schools became increasingly
apparent, given the limited time and resources communities were able to provide. Although
there was almost a gender balance in secondary enrolment overall (girls’ enrolment was 46%
of the total), boys benefited more from the better-resourced state schools while girls were
over-represented in the poorer quality community schools (Rugh and Bossert 1998).
Harambee schools became merged into the government system in the mid 1980s, when all
non-private schools began to receive the same per student government subsidy, although their
structures and facilities remained of poorer quality. Locally-supported community schools at
both the primary and secondary level have also been in existence elsewhere. For example,
urban Zambia as overflow ‘state’ schools, differing from government schools only because
they are completely funded by local contributions and fees (Hyde 2003).
International agency-supported community schools
More recently, community schools have become a way of donors by-passing inefficient
central governments by giving support directly to schools. They often provide their support
through international NGOs, who in turn work with local NGOs. They aim to build on the
strengths of NGOs who have a comparative advantage of working at the grass-roots level and
2
being responsive to local needs. A number of international NGOs are actively involved in
promoting community schools, including Save the Children (UK and US), Care International,
ActionAid, World Education, as well as UNICEF. Community involvement in these schools
can comprise a broad range of forms of participation (see Table 1), including for example:
 School construction and maintenance
 Payment of teacher salaries
 Recruitment of teachers
 Modification of curriculum (including fewer, more relevant subjects, choice of
language etc.).
Two types of international agency-supported community school are discernible: ones
intending to integrate ultimately into the state system, and ones operating as an alternative to
the state system. Those intending to integrate into government system usually operate in a
similar way to state schools, using the same curriculum for example. The involvement of the
community in these schools is more limited, and can be seen as a temporary, stop-gap,
solution in response to the lack of government resources.
Community schools which operate as an alternative to government schools are often modelled
on BRAC schools in Bangladesh, sharing similar features. These have the advantage of being
most flexible and responsive to local needs, and are perhaps the closest to a genuinely
participatory model of community participation, with attempts to engage community
members in a broad range of activities (Table 1). In particular, these community schools aim
to address a range of problems evident in the conventional state schools through, for example:
 Reducing distance between home and school
 Responsiveness to local conditions, including children’s work patterns
 Providing an appropriate curriculum in local languages
 Recruiting teachers familiar with local environment at lower cost (with lower
qualifications), and encourage recruitment of local female teachers
 Shorter pre-service teacher training together with on-going in-service training.
Many of these responses address some of the problems which have been found to
disproportionately affect girls’ educational opportunities in SSA context (Colclough, Rose
and Tembon 2000). These include, for example, the fear for girls’ safety when schools are a
long way from home; the demand for girls’ domestic work; and the lack of female role
models in schools and the community. However, they create a challenge for children to
integrate into the state system after completion of the initial cycle in a community school
given the difference in curriculum. There is a danger, therefore, that a two-tier system is
created with those attending community schools marginalised from the state system,
attendance at higher levels of which is likely to be necessary to gain access to job
opportunities. The most evident example of these alternative community schools in SSA are
those supported by USAID through Save the Children-US first in Mali from 1992 and later in
Malawi in 1995 (termed Village-based schools). As will be seen, there is a very wide
variation of success between community schools in these two countries.
In many SSA countries, a number of community school programmes (both integrated and
alternative models) supported by different NGOs co-exist. These are often small-scale
operating in areas under-served by the state system, although attempts to scale-up coverage of
community schools are evident in some countries.
3
Community participation in government schools
As mentioned, construction of government schools has always been supported by community
contributions in many SSA countries. In Malawi, for example, from the commencement of
formal education, self-help was recognised as important by both the Missions and the colonial
administration, predominantly with the aim of supplementing the insufficient resources
available for education, as well as of making people more involved in their children’s
education. At this time, a deliberate policy of government resources favouring urban areas,
while rural areas were expected to develop education facilities through self-help projects, on
the assumption that self-help was more difficult in urban areas. Prioritisation of public
resources towards urban areas continued post-independence, despite recognition that self-help
projects in rural areas often failed due to limitations on the time and resources of poor
members of these communities. This has contributed to the uneven development of schooling
opportunities which have continuously favoured urban areas (Rose 2002).
Community participation has been continuously promoted and formalised through both
international and national policy, with even greater attention paid to it in recent years. It is not
coincidental that a more explicit emphasis on community participation has corresponded with
the economic crises which have adversely affected education systems in sub-Saharan African
countries since the 1980s, together with rapid expansion of school systems in the context of
the drive for achieving universal primary education and associated abolition of fees to
stimulate demand, necessitating the search for alternative sources of resources (Bray with
Lillis, 1988; Bray, 1996; Shaeffer, 1992; Watt 2001). In practice, community participation in
international and national policy is often at best a form of pseudo-participation, linked with an
attempt to mobilise, and make more efficient use of, resources (see Table 1). The World Bank
(1995) proposes, for example, that, while user fees at the primary level are no longer
supported, cost-sharing with communities is desirable particularly where public resources are
insufficient. This is evident in Malawi where, following the abolition of primary school fees
in 1994, alternative sources of resources for education were required. The 2000 Policy
Investment Framework states, for example:
‘Although local communities and parents are increasingly playing a role in
educational finance especially with regard to sharing in the cost of buildings and their
maintenance, transport to schools, food, uniforms, learning materials and extracurricular activities, their contribution remains unquantified. Community participation
is particularly significant at the primary level. More than 75% of Malawi’s primary
schools have been built with the support of local communities. Primary school
maintenance has largely been a responsibility of communities.’ (Ministry of Education
2000: 4).
The formation of school committees is an important way in which community involvement is
being promoted in many countries, often with stipulation of quotas for females on the
committee.
Community participation in cross-sectoral programmes
Social funds have also become an important way for some donors to divert their funds
directly to the communities they intend to support in recent years. Social funds consider
community participation to be important both for the identification of priority areas as well as
for carrying out programmes. The inter-sectoral Malawi Social Action Fund (MASAF),
supported by the World Bank, has been particularly influential in the education sector since
the 1990s, with up to two-thirds of the first round of MASAF funding for community projects
allocated to primary school construction (Parker and Serrano 2000). The apparent high
4
demand for education projects in MASAF has been attributed to the introduction of Free
Primary Education (introduced in 1994) as a result of which the number of children attending
school increased dramatically and, therefore, meant that more classrooms were required. In
addition, since school committees are already established in most communities, these provide
a structure from which MASAF funding could be applied. Furthermore, headteachers have
played an important role in creating awareness of MASAF within communities, so they are
most aware of the opportunities available (Kishondo 2000). Similarly, 70 percent of projects
arising from the social fund in Zambia were allocated to schools. The reason for the
prioritisation of education was seen to be related predominantly to the extended presence of
PTAs at the local level (Parker and Serrano 2000). Investment funds focusing specifically on
education have been established in Ghana (Condy 1998). Their objective has been more
specifically to improve the teaching and learning environment as a result of increased
community participation and ownership. However, as with social funds in Malawi and
Zambia, evidence from Ghana indicates that they were heavily reliant on a few local teachers,
local elites or external advice and support, given that community members often lacked skills
to conceptualise and plan a project, and the ability to handle finance, budgeting and
implementation.
Social funds have played an important role in placing a value on community contributions. In
Malawi, for example, communities are involved in identification, preparation and financing
with the intention of improving prioritisation and efficiency in the use of resources.
Community co-financing is seen as a way to ensure community ownership and as a true
reflection of demand (World Bank 1996). Thus, the community is expected to contribute 20
percent of resources to the project, with contributions of labour and materials valued at their
market rates. There are potential gender implications of this form of ‘marketisation of
community participation’ (Rose 2003), as discussed below.
In general, these different forms of community support for education range from ‘pseudoparticipation’, most apparent in the models of formalisation of community participation in
social funds and government policy which in practice promote community participation
primarily as a way to extract resources, to community schools which are programmaticallysensitive models with broader aims incorporating attempts to be more genuinely participatory
in a variety of school-related activities (see Table 1). Even so, where community schools are
promoted by international agencies, they still reflect a top-down model of introducing
community participation.
Examples of the experience of community involvement in education in SSA
While community participation in schooling has played an important role in education
systems in SSA, both positively and negatively, this section will focus on its relationship with
gender outcomes in particular. It will first consider community participation as a means to
improving equitable educational outcomes, followed by a review of community participation
in education as an end in itself. With increased international attention on community
participation, several studies have been undertaken to assess the effectiveness of programmes,
particularly evaluations by NGOs and donors involved in supporting the programmes
(although not all of these are publicly available). These programmes are likely to differ from
those which develop without external support, partly due to differences in availability of
resources (Miller-Grandvaux and Yoder 2002). This raises questions about whether
externally-supported schools are more cost-effective than their government counterparts as
often claimed. Tietjen’s paper (1999) is a rare attempt at assessing this, although even this
5
recognises that some costs cannot be fully accounted for. Even so, she concludes that, while
they have a range of benefits, community schools cannot necessarily be considered a cheap
alternative to the state system as often claimed.
Despite the attention given to community participation as well as to gender issues within
education, few studies provide a detailed analysis of the implications of the two issues
combined. In most cases, a brief mention of some aspect of gender is included within the
context of a broader review of community schooling. It is, however, possible to discern some
general trends, as reported below.
Community participation as a means to improving educational outcomes
Access and retention
Several studies report that community schools are successful in improving access to
schooling, and some note that girls’ participation in schooling in particular has improved.
Programmes supported by NGOs in Ethiopia, Ghana, Mali, Guinea, South Sudan and Uganda
are all reported to have had some success in this regard (Miller-Grandvaux and Yoder 2002).
In Ethiopia, for example, total enrolment has increased by 8.9 percent in the region where
World Learning operates a community school programme, and girls’ enrolment has increased
by 13.8 percent. In these schools, girls’ attendance in school also improved (with 36 percent
of girls in class, compared with 28 percent in government schools. The gross enrolment rate
for girls in the district in Guinea where Save the Children is working has increased from 31
percent to 37 percent. In South Sudan, CARE’s work in sensitising communities about the
importance of sending children, in particular girls, to school is reported to have increased
girls’ enrolment by 96 percent. Girls comprise 47 percent of enrolment in ActionAid’s
community schools in Uganda, and almost half of those transferring to government schools
are girls (Miller-Grandvaux and Yoder 2002). In addition, Save the Children-US supported
schools in Uganda are also reported to have benefited girls’ access to education in particular
(Box 2).
These successes in increasing girls’ enrolment are clearly important, although the reasons
behind them generally need further detailed investigation. Given that community schools are
established in areas where schools previously did not exist, children are likely to have greater
access to some form of schooling, and often the choice is between a community school or no
school. It is usually not apparent, however, whether the improved chances of girls’ enrolment
derive specifically from community participation, or would also occur if a state school were
provided in a similar location. It is perhaps more likely that international support for
community schools is useful in providing school places than would otherwise be available,
due to insufficient government resources available to provide an adequate number of state
schools. In Mali, for example, the coverage of villages with schools in two districts increased
from 12 percent to over 40 percent within four years of the programme (Muskin 1999). Since
these new schools are placed in, rather than near, communities they alleviate parents’ fear of
long walks for girls. If state schools had been provided in these areas, there is no reason to
expect that the effect on enrolment would have been any different. In many of the countries in
which Save the Children-US operates, the best location for the school is negotiated with the
community to address this obstacle directly (Dowd and Greer nd). In this case, community
involvement might have a beneficial effect in ensuring the school is appropriately located.
6
Box 2: CHANCE Schools in Uganda
Village Schools help children, especially girls, stay in school
At the government's invitation in 1999, Save the Children initiated a pilot education program with 20
locally constructed schools in Nakasongola District, about 150 kilometers north of Kampala. Called
Child-centered Alternatives for Non-formal, Community-based Education (CHANCE), the Village
School program was envisioned as a high-quality, inexpensive, and sustainable basic education
program with significant community participation and gender parity. It targeted disadvantaged children
who were unable to access the formal schools. Many of these children were from fishing and the seminomadic pastoral communities traditionally left out of the mainstream. CHANCE has since dramatically
changed the education landscape throughout the district and Uganda government officials and others
acknowledge its positive impact on educational access and performance compared to the formal
educational system.
For example:
- CHANCE classes achieved an increase in academic performance in nine months as opposed to 12
months in formal schools, and 3 hours per day as opposed to 5.5 hours per day in the formal schools;
- CHANCE limits class size to 40 children and has a textbook-to-pupil ratio of 3:1; and
- Attendance is almost equally balanced between males, 49 percent, and females, 51 percent.
To date, CHANCE has brought 1,800 marginalized children in the underserved Nakasongola District
into schools set up by Save the Children: 20 pilot classes begun in October 1999 have grown to 48
schools, 73 classes and 2,500 students. About 600 students have completed three years of primaryschool studies.
http://www.savethechildren.org/uganda.shtml
According to Hyde (2003), the programmes that are explicitly based on the BRAC model (in
Save the Children-US community schools in Malawi and Mali) include access and retention
of girls high on their agenda. In both these countries, the evidence suggests that the proportion
of girls attending is higher in these schools compared with state schools. This is, however,
partly due to one of the criteria for establishing a community school in a particular area
stipulating that 50 percent of those enrolled should be girls. Save the Children-US reports that
in Guinea, Mali and Senegal, the gender parity in enrolment criterion is a ‘non-negotiable’
principle. If the community does not agree to classes that are half boys and half girls, the
partnership to establish and manage a new school is placed on hold while additional
negotiating and awareness raising efforts are undertaken. In Mali, school management
committees and parents agree to maintain the 50-50 ratio in the classroom until the end of the
primary school cycle (six years) and parents are subject to fines for removing girls from
school. This, they suggest, shows strong support for girls’ persistence once they are enrolled
(Dowd and Greer nd). However, the reasons for girls’ withdrawal would need to be
investigated to ensure that those households which are already suffering significant hardship
and cannot afford to keep girls in school are not penalised further.
Save the Children-UK also operates the non-negotiable principle of 50 percent girls’
enrolment in their community schools in Mali (which differ from the Save the Children-US
schools in terms of adopting the integrated rather than alternative model). It is noted that
where this challenges traditional assumptions, ‘an element of persuasion comes into the
7
picture’ (Molteno et al 1999: 77). For example, they commission a group of musicians to
perform songs in an attempt to encourage consensus, although a degree of bargaining is often
also needed for villagers to ‘accept’ the messages. It is, however, proposed that the
community does ultimately genuinely support the principle.
In Save the Children-US community schools in Mali, of the 60 children enrolled at a time into
two classes, it is stipulated that 30 should be girls and 30 boys (DeStefano 1996). Soon after
their establishment in 1992, enrolment in 62 community schools in one district indicated that
1,361 girls were enrolled out of a total of 2,718. Gender parity was, therefore, evident by
design. This does, however, compare favourably with the national figures for government
schools, with a total GER of 33 percent, and girls’ GER just 25 percent (DeStefano 1996). A
comparison of eight government schools with 10 community schools in the same district
indicates that dropout rates are extremely low for girls in community schools (0.58 percent,
compared with 2.37 percent in government schools) (Muskin 1999) (see also Box 1). This
suggests that, where appropriate facilities exist and there is external pressure is exerted to
enrol both boys and girls, it is possible for girls to enrol in equal numbers to boys. However,
as argued below, it is unlikely that this outcome is a result of broader transformation of
perceptions of gender roles in the community.
Box 1: Village-based schools in Mali
Villages Most Likely to Succeed
Save the Children’s primary education efforts began in Mali in 1992. At that time, the country’s primary
enrollment rates were among the world’s lowest — under 20 percent in rural areas, and as low as 5
percent for girls. Some villages sent no children to primary school. Save the Children has played a
lead role in the development of village schools in Mali demonstrating: (1) community ownership, (2)
relevant curriculum, (3) absolute gender equity, and (4) teacher training. Each village school is the
result of the efforts of village committees, organized around their need to provide basic education for
their children. To make the schooling relevant, the following components are incorporated into initial
planning and design: (a) the school year is brought into line with the agricultural year; (b) the language
of instruction is the indigenous language of the villagers; and (c) the village school offers a learning
package based on literacy and math to be attained over a three-year cycle. The program also includes
school management training. This school year, 782 village schools are operational with nearly 50,000
students enrolled, 41 percent of whom are girls. In addition, 14 percent of the 1,330 teachers in these
schools are women.
http://www.savethechildren.org/mothers/programs/education.htm
One way of increasing girls’ enrolment in community schools is through awareness-raising
within communities. The methods employed vary between countries. For example, in Malawi
this has consisted of Save the Children-US staff and Ministry partners hold regular meetings
and focused training for key members of the community (including school committee
members) on how to encourage girls to begin and/or continue education, while in Burkina
Faso, village workshops undertake analysis and discussion of gender issues in education
(Dowd and Greer nd). Attendance and retention is enhanced by monitoring and following up
on absence. In Guinea, a PTA member is given the responsibility of monitoring absence,
while in Malawi it is reported that monitoring absence is undertaken as a partnership between
community, teachers and children who follow up with the families of girls who are frequently
absent. This effort includes child-to-child strategies as well as home visits by school or PTA
members, headteachers, teachers or mentor teachers from neighbouring schools. Save the
8
Children and Ministry officials offer training for these participants in guidance and
counselling techniques for use with girls as well as their parents during such visits (Dowd and
Greer nd).
Adaptation of school schedules to take account of the demands for girls and boys work is also
sometimes adopted. This can include providing a reduced form of the national curriculum
focusing on basic literacy and numeracy skills in particular, as is the case in Uganda. In Mali,
the daily and annual school schedule is altered to take work patterns into account, and the
curriculum has been adapted to use local languages, and place greater emphasis on local
knowledge. Enhanced childrearing and health knowledge as well as skills in credit
management and mircoenterprise are reported to have increased the value that families and
girls place upon schooling in these rural communities (Dowd and Greer nd), although this
focus could raise concerns for reinforcing gender roles and responsibilities rather than
attempting to address and transform them.
Attempts are also made in the Save the Children-US community schools to provide female
role models, including by encouraging the recruitment of female teachers from the local area.
In Mali, teachers are recruited from the Save the Children village literacy course or with just a
few years of primary schooling. However, despite efforts to recruit women teachers, only 13
percent are female (Tietjen 1999).
As a result of the general success of the programme, the number of community schools in
Mali has grown significantly to 2,338 (one-third of primary schools in the country) with Save
the Children-US working in partnership with 16 Malian NGOs. As a result, it is reported that
girls’ enrolment has improved significantly, particularly in remote areas (Hyde 2003; Dowd
and Greer nd). This can be attributed to its holistic approach to addressing the underlying
causes for gender inequality in access to schooling, including:
 requiring gender parity in enrolment
 encouraging attendance and retention of girls
 monitoring absence of girls
 awareness raising on the benefits of girls’ education
 adapting the school day around child work, including girls’ domestic chores
 role models (female teachers, school committee members, trainers, NGO staff, and
other education officials and professionals).
However, these interventions are not attempting to transform gender relations, but relate to
existing values towards girls’ and boys’ roles with households and communities. For example,
the school day is adapted around girls’ domestic chores, without questioning whether it is
appropriate for girls to take on the bulk of this work, rather than promoting the view that the
work should be more equally distributed between boys and girls. Whether it is realistic to
expect community participation to result in such transformation is, in any case, debatable.
Despite its success in Mali, it is not necessarily possible or appropriate to replicate this
positive experience in other areas of SSA. A similar model was also established by Save the
Children-US in Malawi, but was less successful due to the lack of support and flexibility of
the government. In particular, the government was concerned about the effect on quality of
education of recruiting untrained teachers who had not completed the secondary school cycle.
As a result, these community schools did not progress beyond a pilot (Hyde 2003). Hyde
notes that there has been a clear difference in the rate of expansion of community schools
between Anglophone and Francophone countries, an important reason for which is the extent
9
to which the state plays a role in sanctioning the opening of new community schools, with the
state playing a more active role in this regard in Malawi (and Uganda) than in Mali (and
Chad).
Box 2: Community participation in government primary schools in Senegal
Boosting Community Involvement in Primary Schools
Gouye Gui (meaning 'Baobab tree' in the local language) is a rural community located 40 miles from
the regional capital of Diourbel and 120 miles from Dakar. This region has some of the lowest
enrollment rates in Senegal. Severe poverty is a deterrent for schooling as illiterate communities can
afford neither health nor education services. Girls are further hampered by conservative cultural and
religious traditions, including early marriage for girls (by their mid-teens). In 2000-2001, the village
primary school had 150 students with just 12 girls in first grade. Low attendance in school, combined
with a lack of basic supplies and an unhealthy school environment made schooling a low priority for
poor families.
In April 2001, the Education for Development and Democracy Initiative (EDDI) in Senegal launched a
community sensitization campaign focusing on girls' education as well as training in overall human
rights, literacy, and life skills. Two months later a School Management Committee (SMC) was elected
with more than 50% women members. The SMC received training in leadership, budgeting, and
management, and the skills thus developed have enabled them to first maintain an inventory of school
supplies and successfully manage and complete a school renovation project. The renovations have
included the construction of two sets of latrines (one for boys and one for girls), rehabilitation of the
three classrooms, as well as construction of a protective wall surrounding the school. Finally, in
December 2001, teachers received training in gender awareness pedagogy.
The effects of this integrated system of interventions for the academic year 2001-2002 have been
many. 27 girls are now enrolled in first grade (an increase of 125%), and the total of 150 students is
now divided evenly between boys and girls. Several teachers interviewed stressed the importance of
subsidized school supplies to increase the likelihood of the most vulnerable students (including girls
and the very poor) to attend school, as well as in improving daily attendance rates of those already in
school. Protective walls, renovated classrooms, latrines and water points have improved the security
and hygiene of the school environment, attracting interest from the community and providing
opportunities to apply the lessons learned on environment, health and resource management.
Teachers also proudly point out their new class duty rosters, on which chores are now evenly
distributed between boys and girls - a significant break from the tradition of having girls sweep and
clean while boys play! The SMC has even played a role in arbitrating a long-running dispute within the
village regarding a grain mill, the result being that the village cooperative now has an incomegenerating project and the workload of women is reduced. The Gouye Gui school community, through
the SMC, recently made an independent request for adult literacy classes, showing that their new
skills are enabling them to initiate and organize their own development activities.
http://www.usaid.gov/regions/afr/ss02/senegal4.html
Attempts to promote girls’ enrolment in government schools through community initiatives
can also result in success for access, as the example of Senegal illustrates (Box 2). As with the
Mali community schools, the success of this example can be partly attributed to a holistic
approach being undertaken to address the range of identified problems. Where gender issues
are not systematically integrated into the programmes or appropriately monitored, they are
more likely to fail. In Malawi, DFID-supported primary community schools initially
considered the need to involve women in school committees, but did not seek to investigate
and address constraints affecting girls’ under-enrolment and persistence in remote areas. This
was later rectified, although a lack of clarity about where responsibilities for monitoring and
meeting gender-related targets was still evident (Al-Samarrai, Bennell and Colclough 2002).
10
Performance and transition
More limited evidence is available on the relative performance of pupils by gender in
community and government schools in SSA. A comprehensive review of community schools
in SSA suggests that performance is higher in community schools compared with government
schools, although information is often not disaggregated by gender (Miller-Grandvaux and
Yoder 2002). In Malawi, children in Save the Children-US community schools are reported to
have scored 30 percent higher than their government school counterparts, and girls in these
schools learned significantly more compared with their counterparts in government schools.
Similarly, students in Save the Children-US CHANCE schools in Uganda are reported to
perform significantly better than those in government schools, even though they are only in
school for 9 months, rather than 12 months, and have a shorter school day (Box 2). Reasons
for the improvement in performance include ones both directly and indirectly related to
community management, as well as due to additional resources available from external
sources which support the reduced curriculum, improved teaching methods and smaller class
size. Furthermore, the results do not control for family background or cognitive ability. Thus,
attribution of differences in performance to community involvement needs to be treated with
caution.
There is some ambiguity in the evidence on performance between government and
community schools in Mali, where more detailed analysis is available. Muskin (1999)
suggests that performance of both boys and girls is, in general higher in community schools
compared with their government school counterparts (Figure 1). However, a gender gap
remains, despite attempts made to make sure that the learning environment was not gendered.
Improvement in performance is most evident in language tests, probably as a result of the use
of local languages in community schools, and a narrowing of the gender gap is also apparent.
Girls’ performance also improved slightly more than boys in mathematics in community
schools. Surprisingly, both boys and girls in community schools performed lower than their
government school counterparts in the local knowledge test, even though this was an explicit
focus of the community school curriculum. Significant differences between community and
government schools are evident in performance in language and local knowledge tests (but
not mathematics) controlling for family background (including schooling level of other
members of the household, weighted index of household possessions, size of landholding and
receipt of remittances). On the basis of multivariate analysis, the author concludes that:
‘Overall, the regression outcomes demonstrate that while other factors associated with
some individual schools...obviously influence student outcomes for both types of
school and on all three tests, the strongest implication is that the type of school clearly
bears the greatest effect on a student’s learning’ (Muskin 1999: 53).
While this analysis is more rigorous than most other studies available comparing performance
in government and community schools in SSA, it still does not control for resource inputs
(affecting class size, for example) which, as discussed, are likely to be greater in the
internationally-supported community schools. As such, the results are not conclusive as
differences in performance could actually be due to additional inputs being available from
international sources in community schools.
11
Figure 1: Performance in government and community schools by gender
90
80
70
60
50
40
30
20
10
0
language
maths
government - boys
community - boys
local knowledge
government - girls
community - girls
Source: Muskin 1999.
In contrast to Muskin’s results, more recent evidence by Sangara and Diarra (2002) (based on
different tests) indicates that the average score in mathematics is slightly higher in
government schools compared with community schools (97 percent and 93 percent,
respectively), while the opposite is the case for the general knowledge test (97 percent and 99
percent respectively). Unfortunately, these results were not disaggregated by gender, and do
not control for family background or availability of resources. Similarly, the pass rate for
community schools is reported to be lower than for government schools in the end-of-primary
school exams, with a pass rate of 32 percent and 45 percent, respectively, in 2000-01 (MillerGrandvaux and Yoder 2002).
More limited evidence is available on transition between community schools and the state
system. One study notes, however, that 19 percent of those enrolled in ActionAid-Uganda
ACCESS schools transferred into government schools at the end of the four years, and more
than half of them were in the top half of their classes by the end of the first year in
government schools (Wrightson 2001, cited in Miller-Grandvaux and Yoder 2002). This
suggests that a large number of those who attend these community schools do not have the
opportunity to continue with their schooling (and it is likely that those who do proceed are
amongst the highest achievers).
Community participation as empowerment – or entrenchment of gender relations?
Many studies analyse the ‘community’ as if it were a homogenous group of people, devoid of
power relations within it, implying a vision of ‘community’ as a network of shared interests
and concerns. In reality, however, a community is unlikely to be a homogenous group with a
common voice and shared set of views. By emphasising common knowledge, the promotion
of community participation can fail to acknowledge the ways in which local power is
reinforced (Wolf et al, 1997; Mosse, 2001). This is particularly important for considering the
12
extent to which gender dynamics are influenced by innovations seeking to promote
community participation.
With respect to community participation in existing government schools, a recent survey in
Malawi indicates gender inequalities in community contributions within households. Of 238
households interviewed, 70 percent of those involved in providing non-monetary
contributions were women (Rose 2003) Wives of heads provided most of the labour, followed
by female heads of households, with male heads least likely to contribute their labour.
Discussions with parents and school committees generally reinforced the view that the burden
of community activities was placed more on women. A gender division of labour was also
apparent, reflecting responsibilities undertaken by men and women in society more generally.
At one school visited, parents mentioned, for example, that women fetch water while men
mix the soil and water and take the bricks to the builders (Rose 2003). Similarly, a previous
study by the Government of Malawi/UNICEF (1993) reports that while women usually get
involved in community activities, such as maintaining water supplies, school construction,
and child care and literacy schemes, men take on more of the community leadership roles.
Kadzamira and Ndalama (1997) also note that, in the areas visited for their study, women
carried out the bulk of the work and roles were usually assigned according to culturally
accepted gender roles. Women were usually responsible for hauling sand and water for
construction or moulding bricks, while men were responsible for moulding bricks, brick
laying, carpentry work and other skilled jobs.
Although innovative programmes aimed at encouraging community involvement in schooling
address a range of constraints faced by children from poor households, they can potentially
increase the direct costs that poor households face. Community schools are often designed to
be established in poor, remote communities, but can require greater contributions from the
community than government schools in wealthier areas in terms of fees, support for
instructors, and school construction (Hyde 2003). In Mali, for example, the Save the
Children-US community schools require school management committees to mobilise
resources for school construction, and subsequently to collect money from the community for
teacher salaries (Tietjen 1999). Although the cost of construction and level of teacher salaries
overall are lower than in government schools, the resources in these schools are provided
directly by the community rather than from the government as is the case in the higher cost
schools which are established in less remote areas. The shift from household to community
contributions which has occurred following the abolition of fees and subsequent promotion of
community participation in some countries, has potential implications for the relative burden
placed on women and men in the household. In Malawi, men were most often responsible for
paying fees before their abolition while, as noted, women provide more community support
(Rose 2003). Similarly, in Benin, women have been found to be involved in providing human
contributions, while men contribute more in financial and material terms (Salami and
Kpamegan 2002).
In recognition of the burden that participation can place on members of the community, some
programmes attempt to allocate a monetary value. However, marketisation of community
participation has, if anything, intensified intra-household inequalities in community
participation. For example, at a school visited in Malawi, it was noted that making of bricks
undertaken by men was paid since it required skill, whereas carrying water by women, often
for long distances, was not given any value despite the amount of time and energy involved.
Furthermore, those with skills who, in any case are likely to be better off members of the
community, were paid while those less likely to have any access to paid employment were
13
expected to provide their services voluntarily (Rose 2003). Kadzamira and Ndalama (1997)
also found that men participated more in services for which they were paid in Malawi,
whereas all contributions provided by women were free. DFID-supported primary community
schools in Malawi engaged local contractors who would be more likely to employ labour
from within the school area, so that financial benefits would remain at the local level. Female
contractors were encouraged, with 25 percent of contractors on the database being female in
1999. Nine women of 75 contractors awarded a contract, and of the 14 contractors awarded a
second contract, nine were women (Durston and Nashire (2001). Despite this, Mkamanaga’s
study (1998) of the community schools indicated that only 20 percent of the paid labour force
involved in the community projects were female. Of those who were paid, men would earn
more because they were involved in various tasks whereas women were mostly given what
were considered to be ‘lighter jobs’ such as drawing water which were given lower value.
In terms of the composition of school committees, attempts have been made to ensure
diversity of their membership in particular to include women in a number of countries in SSA.
However, even these generally work within existing community structures with their inherent
power relations, rather than attempting to challenge and transform them. There is no attempt,
for example, to ensure that women feel free to speak out on these committees, and have an
equal voice in decisions being made. In Malawi, legislation stipulates that one-third of places
on school committees should be reserved for females. Although attempts at improving gender
equity in decision-making at the school level are evident, in practice the involvement of
women is usually at best limited to giving them a place on the committee, rather than ensuring
their active participation. Furthermore, in reality women continue to be outnumbered on the
committees as, in many cases, the quota is not met and rarely exceeded. At seven of 20
schools visited there were no women on the committee (Rose 2003). At one school, it was
noted that the lack of women on the school committee was because, although they had been
selected, they refused to join the committee because they said that they have too much work
to do in their village. Furthermore, it was evident from discussions held with school
committees that women either did not turn up to the meetings or, if present, would often not
speak.
Similarly, in Ghana it was found that school committees rarely meet national requirements for
women representatives, since membership is based on existing positions of leadership in the
community and school which are usually held by men (Condy 1998; Pryor and Ampiah
2003). When meetings are held, women often cannot attend because the timing was
inappropriate. Women generally contributed communal labour by bringing water, bricks, sand
and mortar to schools, but many were not sure what it was used for. However, women
suggested that they would discuss their ideas with their husbands which would be passed on
to the committee. For example, women in one community mentioned that they had the
initiative of encouraging food sellers to sell their food close to the school to provide food for
their children so they could stay in school (Condy 1998). Evidence from Uganda also
indicates that discussions in committees are mainly dominated by men, with only a few
women speaking. Women feel inhibited from taking part in formal discussions where
traditions assign decision-making roles to men (Suzuki forthcoming).
Furthermore, a disjuncture between those deciding on contributions to be made and those
providing most of the non-monetary support to schools was evident. At one school visited in
Malawi which was preparing to build a new classroom block, the school committee divided
up the work to be undertaken amongst the various villages within the school catchment area.
It was decided that the bricks should be made in one of the villages furthest away from the
14
school. While the job of making the bricks was men’s work, carrying them to school over a
long distance and difficult terrain was the responsibility of the women (children from that
village mentioned that they had problems coming to school because they had to walk through
dense wooded hills where snakes were abundant). Women’s lack of influence within the
decision-making body resulted in this unfortunate outcome. Although there was a female
member of the school committee, she had been silent throughout the discussions (Rose 2003).
In situations where children from the poorest households are least likely to be in school in the
first place and contributions are expected from all community members, the poorest, who are
unable to afford other direct and indirect costs of schooling, are in effect subsidising
schooling costs of the better-off. In addition, the burden of community contributions is likely
to depend on family structures, and be altered as these change. Where children have lost their
parents as a result of AIDS, they are often less likely to be in school. This is particularly the
case for girls who have to care for sick relatives, or take on the domestic responsibilities of
mothers who have to earn an income to support the family. In addition, children who have lost
their parents are often living with grandparents who may be too old to be actively involved in
schooling. Furthermore, it is possible in countries in SSA where the AIDS pandemic is
spreading fast, that the next generation of children may have neither parents nor grandparents
alive. Community contributions are likely to become increasingly burdensome in such
circumstances, and are likely to increase inequalities between and within communities.
Conclusion
In conclusion, it is evident that a number of initiatives involving the community are taking
place, often with the support of international donors and NGOs some of which are modelled
on examples from South Asian experience. Evidence based on rigorous analysis is patchy, but
suggests at least that schools now exist in remote areas where children previously did not
have access to schooling opportunities, and that improvements in the gender composition of
enrolment in these schools are evident partly due to the reduced distance between home and
school, as well as to other initiatives aimed at encouraging girls’ enrolment in school. Based
on the evidence available, it is not, however, possible to determine whether these
improvements are attributable to community participation per se, or whether state schools in
the same location and with the same resources could result in an identical outcome.
Furthermore, concerns are raised about whether community schools established as
alternatives to the state system are creating a two-tier system, with limited opportunities for
children attending these schools to continue with their education in the state system. As an
end in itself, community participation in schooling appears to have resulted in an
entrenchment and reinforcement of gender relations, rather than empowering those
traditionally excluded from more genuine aspects of participation. Referring to the matrix
presented earlier (Table 1), evidence suggests that, of the range of forms that participation can
take, in practice it is often mainly confined to contribution of resources for building and
maintenance. However, some community-school models do attempt to involve communities
in a broader range of activities in a genuinely-participatory way. Even in these examples,
genuine participation in decision-making is most frequently undertaken by men in the
community, while women’s role is confined to a pseudo-participatory one, in terms of
providing their unpaid time for school construction activities, for example. However, it is
possible that improvements in access and performance in schooling by gender as a result of
greater community involvement in schooling could have an impact on gender relations in the
longer term, if this is genuinely desired and addressed.
15
References
Al-Samarrai, S., P. Bennell and C. Colclough (2002) ‘From projects to SWAPs: An
evaluation of British aid to primary schooling 1988-2001’ DFID Evaluation Report
EV639
Bray, M. (1996) Decentralization of Education: Community Financing, Washington DC:
World Bank
Bray, M. (2000) Community Partnerships in Education: Dimensions, Variations and
Implications. EFA Thematic Study, Dakar
Bray, M. with K. Lillis (1988) Community Financing of Education. Issues and Policy
Implications in Less Developed Countries, Oxford: Pergamon Press
CARE International Ghana (2003) ‘Reaching underserved populations with basic education in
deprived areas of Ghana: Emerging best practices’ Accra: CARE International
Colclough, C., P. Rose and M. Tembon (2000) ‘Gender inequalities in primary schooling: The
roles of poverty and adverse cultural practice’ International Journal of Educational
Development Vol 20 No 1: 5-29
Condy, A. (1998) 'Improving the quality of teaching and learning through community
participation: Achievements, limitations and risks. Early lessons from the School
Improvement Fund in Ghana', DFID Social Development Division Working Paper No 2,
London: DFID
DeStefano, S. (1996) ‘Community-based primary education: Lessons learned from the Basic
Education Expansion Project (BEEP) in Mali’ Washington DC: Human Resources and
Democracy Division, USAID
Dowd, A. and H. Greer (nd) ‘Girls’ education: Strong beginnings. Community approaches to
access and quality’ Save the Children-US. http://www.savethechildren.org/girlseducation/
[accessed 8/5/2003]
Durston, S. and N. Nashire (2001) Rethinking Poverty and Education: Attempt by an
Education Programme in Malawi to have an Impact on Poverty. Compare. Vol 31, No 1:
75-91
Government of Malawi and UNICEF (1993) 'Situation analysis of poverty in Malawi',
Lilongwe: Government of Malawi/UNICEF
Hyde, K. (2003) ‘Expanding Educational Opportunities at Primary Level: Are Community
Schools the Answer?’ Mimeo
Kadzamira, E. and M. Ndalama (1997) ‘Primary Community Schools Project. Baseline study
for Southern Region: Consolidated final report.’ Lilongwe: DFID, Primary Community
Schools Project
Kishondo, P. (2000) 'Community project funding in Malawi under the Malawi Social Action
Fund (MASAF) demand-driven approach: Potential for perpetuating imbalances in
development', Journal of Social Development in Africa, Vol 15 No 1: 5-14
Miller-Grandvaux, Y. and K. Yoder (2002) ‘A literature review of community schools in
Africa’ Support for Analysis and Research in Africa (SARA) Project, Washington DC:
USAID
16
Ministry of Education (2000) Policy Investment Framework (PIF). Ministry of Education,
Sports and Culture, Lilongwe
Mkamanga, G. (1998) ‘Primary Community Schools Project: A socio-economic study of the
construction component.’ Lilongwe: DFID
Molteno, M., K. Ogadhoh, E. Cain and B. Crumpton (1999) ‘Towards responsive schools:
Supporting better schooling for disadvantaged children: Case studies from Save the
Children’ DFID Knowledge and Research Papers No. 38. London: DFID
Mosse, D. (2001) 'People's knowledge', participation and patronage: Operations and
representations in rural development.’ In Participation: The New Tyranny? (edited by B.
Cooke and U. Kothari), London: Zed Press
Muskin, J. (1999) ‘Including local priorities to access school quality: The case of Save the
Children Community Schools in Mali’ Comparative Education Review Vol 43 No 1: 3663
Parker, A. and R. Serrano (2000) ‘Promoting good local governance through social funds and
decentralization.’ Washington DC: World Bank
Pryor, J. and J. Ampiah (2003) ‘Understandings of education in an African village: The
impact of information and communication technologies’. Report on DFID Research
Project.
Rose, P. (2003) ‘Community participation in school policy and practice in Malawi: Balancing
local knowledge, national policies and international agency priorities’ Compare Vol 33
No 1: 47-64
Rose, P. (2002) ‘Cost-sharing in Malawian primary schooling: From the Washington to the
post-Washington consensus’ Unpublished DPhil thesis: University of Sussex
Rugh, A., H. Bossert (1998) Involving Communities: Participation in the Delivery of
Education Programs, Washington DC: USAID
Salami N. and G. Kpamegan (2002) ‘Benin’ in A Transnational View of Basic Education:
Issues of Access, Quality and Community Participation in West and Central Africa,
Washington DC: USAID
Sangara S. and I. Diarra (2002) ‘Mali’ in A Transnational View of Basic Education: Issues of
Access, Quality and Community Participation in West and Central Africa, Washington
DC: USAID
Shaeffer, S. (1992) 'Collaborating for educational change: The role of parents and the
community in school improvement', International Journal of Educational Development,
Vol 12 No 4: 277-295
Stiglitz, J. (1997) ‘An agenda for development for the Twenty-First Century’, Ninth Annual
Bank Conference on Development Economics, Washington DC: World Bank
Suzuki, I. (forthcoming) Parental Participation in School Governance in Uganda: who is
responsible for educating children? Brighton: University of Sussex
Tietjen, K. (1999) ‘Community schools in Mali: A comparative cost study’ Washington DC:
USAID
Watt, P. (2001) ‘Community support for basic education in sub-Saharan Africa’ Africa
Region Human Development Working Paper Series. Washington DC: World Bank
17
Wolf, J. et al. (1997) Planning for Community Participation in Education, Washington DC:
Office of Sustainable Development. USAID
World Bank (1995b) Priorities and Strategies for Education: A World Bank Review,
Washington DC: World Bank
World Bank (1996) 'Malawi Social Action Fund Project', Staff Appraisal Report No 15345,
Washington DC: World Bank.
18