In recent years, interest in gender and education has risen because of the 'Education for All' targets and the Millennium Development Goals on gender equality in education by 2015. While the focus has mainly been on access, achievement is also an important aspect. For instance, girls' poor performance reduces their chances of entry into post-basic education in countries like Malawi where selection is on merit.
Central to discussions about gender equality is the notion of quality, which is seen as crucial in expanding girls' capabilities, and accelerating progress towards the international goals (see, for example, Aikman and Unterhalter, 2005; Unterhalter, 2005). This is why it matters that education in Malawi be of good quality, as evidence to date indicates that the quality of most private schools is poor, due to lack of government regulation (Kiernan et al., 2000; British Council, 2001; Staff Development Institute, 2001; Chimombo et al., 2004; Lewin and Sayed, 2005; Rose, 2005; Chimombo, 2009). There are also quality issues in public schools, especially in community day secondary schools (see African Development Fund, 2001; Tudor-Craig, 2002; World Bank, 2004). Over and above, research fi ndings suggest secondary school girls' achievement is considerably lower than that of boys (Kadzamira, 1987, 1988; Khembo, 1991, 1998; Hyde, 1993, 1999; Mbano, 2003). Hence, the paper compares students' achievement in private and conventional public schools in order to arrive at an understanding of which type of school fosters girls' success and why.
The paper fi rst discusses issues around gender quality and the private/ public school divide before giving an overview of Malawi's secondary education. Thereafter, I explain the methods used for data collection and present fi ndings.
Gender, quality and private/public schooling
At the centre of recent debates about quality of schooling is the privatepublic school divide. While Cox and Jimenez (1990), Jimenez et al. (1991), Lockheed and Zhao (1993) and Jacob et al. (2008) argue that private schools are more effective than public ones in enhancing achievement in Colombia and Tanzania, Dominican Republic, the Philippines, and Uganda respectively, the opposite has been found to be true for Tanzania (Chediel et al., 2000; Lassibille et al., 2000; Lassibille and Tan 2001), Rwanda (World Bank 2003) and Indonesia (Newhouse and Beegle 2005). Psacharopoulos (1987) and Lockheed and Bruns (1990) report mixed fi ndings for Colombia and Tanzania, and Brazil, with respect to subject, suggesting that fi ndings differ depending on the kind of schools in a particular context.
According to Riddell (1993), the relative advantages of private and public schools have to be viewed against particular goals. That is, whether a particular school type is cost effective, promotes choice, or increases access. To these, one could add gender equality, as evidence indicates that this is a key issue internationally, let alone for Malawi. However, to date there is little critical analysis of the gender implications of private schooling in developing countries. For example, there is no consensus that the schools promote gender equity in terms of access and most of the work relates to the primary sector (see, Vavrus, 2002; Mehrotra and Panchamukhi, 2006; Tooley and Dixon, 2006; Srivastava, 2006). Where reference to achievement is made, it is not in great detail (Chimombo et al., 2004; Lewin and Sayed, 2005; Chimombo, 2009) to enable one to conclude that the schools enhance girls' achievement. It is therefore crucial to ascertain the relationship between gender differences in achievement and type of school.
Secondary education in Malawi
Secondary education is offered through government schools, grant-aided schools; partly funded by missions, community day schools (formerly distance learning centres operating in buildings supplied by communities), and private schools owned by individuals and some organisations. Government and grant-aided schools are also known as conventional schools, though government provides fi nancial support, in varying degrees, to all the schools, except private ones (see Kiernan et al., 2000; Chimombo et al., 2004; Lewin and Sayed, 2005; Rose, 2005; Chimombo, 2009) for information on different schooling types and fee structures). As Kiernan et al. (2000) point out, there are variations in how different educational systems conceptualise 'private' and 'public' schools. In Malawi, some schools, which are designated as public and are funded by government, are actually owned by religious or local communities.
Private and public schools operate different selection criteria. Students get selected to government schools on merit, with the highest ranked being selected to national schools depending on the district quota, which is determined by the number of candidates in a particular district in relation to the national total, followed by those selected to district (usually boarding), day and then community day secondary schools. Access to community schools was open before their conversion from distance education centres. On the other hand, most private schools take students who can afford fees, though a few administer entrance examinations.
To map broad trends in achievement by gender and type of school, statistics were obtained from the Malawi National Examinations Board from 1997, the year they started being computerised. However, the data could not be disaggregated further into boarding/day, rural/urban, and high/middle/ low fee schools due to time constraints, as it took the Board almost 18 months to organize the data in its current form due to technical problems. While evidence beyond statistics is crucial in making robust judgements about type of school, they begin to point to whether or not type of school makes a difference to academic success broadly and to girls' achievement in particular.
Generally, schools vary in terms of quality. The analysis here is based on conventional public and registered private schools which are mainly those in high and mid-fee range, as most of the low-fee ones do not meet the requirements to operate as examination centres. It has been suggested that the achievement of pupils in these schools is comparable (Chimombo et al., 2004; Chimombo, 2009).
The gender gap and girls' achievement by type of school: a statistical analysis
It is useful to start with gender differences in access to different school types, as this helps to indicate the way in which selection might infl uence achievement. Table 1 shows enrolment fi gures. An important point to note is that the overall gender gap in access is smaller in private than in conventional public schools. Similarly, the gender gap of those who sat examinations is considerably narrower in private than in public schools (Table 2 and Figure 1). Presumably, there is selection bias of those who attend private schools such that the socio-economic background of boys and girls in the schools might be more similar and this can have implications for achievement.
Also worth noting is that the pass rate for both boys and girls is higher in public than in private schools. The pass rate for girls in private schools is below that of girls in public schools, and is considerably lower than that of boys in both types of school. The difference in pass rates between boys and girls has been generally higher for private than for public schools, with the exception of 1998, 2001, 2002 and 2006. Since the pass rates for public school boys for these years are higher than those of private school boys, it could be argued that the gender gap for private schools appears to be narrower only because boys in the schools do not perform as well as those in public schools, but not because girls in the schools are doing well compared to those in public schools.
Turning to girls, Table 3 and Figure 2 point to previous fi ndings that girls perform better in single-sex than in co-educational schools (Hyde, 1999; Khembo, 1991, 1998), though there is no evidence to suggest that girls who get selected to the former schools have better scores in the primary school leaving examination. Table 3 shows that there are considerably more girls sitting examinations in mixed-sex than in single-sex schools, and that the numbers in the former have been rising, while remaining similar in the latter. This could mean that single-sex schools are more selective, or that they do not have the capacity to take in more students, since most of them have boarding facilities. A recent government policy requiring that half of places in its day coeducational schools be reserved for girls partly explains the increment.
It is evident from Figure 2 that girls' achievement is highest in public single-sex schools followed by private single-sex schools then public mixed schools, with private mixed schools being last. Nonetheless, Table 3 demonstrates that in both single-sex and coeducational settings, girls generally do better in public than in private schools, with the exception of 2001 when private single-sex girls slightly outperformed their peers in public girls' only schools. The achievement gap for girls in private and public schools has actually been widening between 2003 and 2006 in both single-sex and mixed-sex schools. So while girls' achievement is problematic, on the whole, they seem to be doing better in public than in private schools.
Conclusion and educational implications
The question of the contribution of different school types towards achievement and achievement gaps is an important one because of its implications for equity between different social groups, and equity is a central concern of 'Education for All' and Millennium Development Goals. This paper has examined differences in achievement by type of school, with particular reference to gender, and indicates that not only do boys and girls do better in public than in private schools, but also that girls do less well than boys in both settings.
These fi ndings are similar to those of Tanzania, Rwanda and Indonesia as noted above. Unfortunately, no study in Malawi demonstrates precisely the relationship between type of school and achievement though private and public schools have been said to differ in terms of student composition, teacher quality, resource/facilities availability and school management (see, Kiernan et al., 2000; Staff Development Institute, 2001; Chimombo et al., 2004; Lewin and Sayed, 2005, Chimombo, 2009) mostly in favour of public schools. These factors are among those that have been consistently associated with high student achievement by studies on the relative effi ciency of private and public schools. Arguably, the elements should affect both boys and girls, but girls seem to be at a greater disadvantage for some reason. Thus, there is a need for further research in order to understand the differentials.
Perhaps, an interesting question that the fi ndings raise is why do families in Malawi choose to use private schools. According to Chimombo (2009), this is because there are no other options as places in government schools are limited. While recognizing that the private sector in the country is highly differentiated in terms of elite versus low quality schools (Chimombo et al., 2004; Lewin and Sayed, 2005), it is important to bear in mind that most of the designated schools follow a European syllabus and administer international examinations, which may help to explain the public school advantage. At the same time, most of the unregistered private schools do not operate as examination centres, which means that the performance of private schools presented here could be an overestimation given that there are more of these schools than the designated schools. Thus, there is a need to re-examine policy to promote schools with the best achievement gains, so that general underperformance might be redressed, especially for girls much as there is also a need to allow private schools to expand in order to meet demand.
African Development Fund (2001), 'Support to community day secondary schools (Education IV project)' appraisal report, Tunis: African Development Bank.
Aikman, S., and Unterhalter, E. (2005), 'Introduction', in S. Aikman and E. Unterhalter, Beyond access: transforming policy and practice for gender equality in education, Oxford: Oxfam, pp. 1-12.
British Council (2001), Quality British Council and value in private school education in Malawi, Lilongwe: British Council.
Chediel, R. W., Sekwao, N., and Kirumba, P. L. (2000), Private and community schools in Tanzania (mainland), Paris: International Institute for Educational Planning.
Chimombo, J. (2009), 'Expanding post-primary education in Malawi: are private schools the answer?' Compare, 39 (2), 167-84.
Chimombo, J., Dzimadzi, C., Kunje, D., and Lewin, M. (2004), The development of private sector providers of educational services in Malawi: panacea or problem? Zomba: Centre for Educational Research and Training.
Cox, D., and Jimenez, E. (1990), 'The relative effectiveness of private and public schools: evidence from two developing countries', Journal of Development Economics, 34 (1), 99-121.
Education Management Information System (2007), Education Statistics, Lilongwe: Ministry of Education.
Hyde, K. (1993), Gender streaming as a strategy for improving girls' academic performance: evidence from Malawi, Zomba: Centre for Social Research, University of Malawi.
_____ (1999), 'Barriers to educational opportunity in Malawi', in S. Erskine and M. Wilson, Gender issues in international education: beyond policy and practice, New York: Falmer Press, pp. 131-48.
Jacob, W. J., Holsinger, D. B., and Mugimu, C. B. (2008), 'Private secondary education in Uganda: implications for planning', Teachers College Record, 110 (4), 867-93.
Jimenez, E., Lockheed, M., and Paqueo, V. (1991), 'The relative effi ciency of private and public schools in developing countries', World Bank Research Observer, 6 (2), 205-18.
Kadzamira, E. C. (1987), Sex differences in the performance of candidates in MSCE mathematics and science subjects, 1982-1986, Zomba: Malawi National Examinations Board.
_____ (1988), Sex differences in the performance of candidates in languages and humanities subjects at MSCE level, 1982-1986, Zomba: Malawi National Examinations Board.
Khembo, D. (1991), 'Comparison of examination results over time', unpublished M.A. thesis, University of London.
_____ (1998), Multiple-choice examinations and sex differences at Junior Certifi cate of Education level, Zomba: Malawi National Examinations Board.
Kiernan, M., Latham, M., Macrae, M., and Read, A. (2000), The development of private secondary education in Malawi: towards a public-private partnership, Lilongwe: CFBT and DANIDA.
Lassibille, G., and Tan, J. (2001), 'Are private schools more effi cient than public schools? Evidence from Tanzania', Education Economics 9 (2), 145-72.
Lassibille, G., Tan, J., and Sumra, S. (2000), 'Expansion of private secondary education: lessons from recent experience in Tanzania', Comparative Education Review, 44 (1), 1-28.
Lewin, K. M., and Sayed, Y. (2005), Non-government secondary schooling in sub-Saharan Africa: exploring the evidence in South Africa and Malawi, London: DFID.
Lockheed, M. E., and Bruns, B. (1990), 'School effects on achievement in secondary mathematics and Portuguese in Brazil', World Bank Policy, Research, and External Affairs Working Paper 525, Washington DC: World Bank.
Lockheed, M. E., and Jimenez, E. (1994), Public and private secondary schools in developing countries: what are the differences and why do they persist? Washington DC: World Bank.
Lockheed, M. E., and Zhao, Q. (1993), 'The empty opportunity: local control and secondary school achievement in the Philippines', International Journal of Educational Development 13 (1), 45-62.
Mbano, N. (2003), 'The effects of a cognitive acceleration intervention programme on the performance of secondary school pupils in Malawi', International Journal of Science Education, 25 (1), 71-87.
Mehrotra, S., and Panchamukhi, R. P. (2006), 'Private provision of elementary education in India: fi ndings of a survey in eight states', Compare, 36 (4), 421-42.
Newhouse, D., and Beegle, K. (2005), The effect of school type on academic achievement: evidence from Indonesia, Washington DC: World Bank.
Psacharopoulos, G. (1987), 'Public versus private schools in developing countries: evidence from Colombia and Tanzania', International Journal of Educational Development, 7 (1), 59-67.
Riddell, A. R. (1993), 'The evidence of public/private educational trade-offs in developing countries', International Journal of Educational Development, 13 (4), 373-86.
Rose, P. (2005), 'Privatisation and decentralisation of schooling in Malawi: default or design? Compare, 35 (2), 153-65.
Srivastava, P. (2006), 'Private schooling and mental models about girls' schooling in India', Compare, 36 (4), 497-514.
Staff Development Institute (2001), Strategic plan for private schools association, Mpemba: Staff Development Institute.
Tooley, J. and Dixon, P. (2006), '"De facto" privatisation of education and the poor: implications of a study from sub-Saharan Africa and India', Compare, 36 (4), 443-62.
Tudor-Craig, M. (2002), Teacher talking time: a policy research report on Malawian teachers' attitudes to their own profession, London: Voluntary Service Overseas.
Unterhalter, E. (2005), 'Global inequality, capabilities, social justice: the millennium development goal for gender equality in education', International Journal of Educational Development, 25 (2), 111-22.
Vavrus, F. (2002), 'Making distinctions: privatisation and the (un)educated girl on Mount Kilimanjaro, Tanzania', International Journal of Educational Development, 22 (5), 527-47.
World Bank (2003), Education in Rwanda: rebalancing resources to accelerate postconfl ict development and poverty reduction, Washington DC: World Bank.
_____ (2004), Cost, fi nancing and school effectiveness of education in Malawi: a future of limited choices and endless opportunities, Washington DC: World Bank.
Martha Kamwendo University of Sheffield
Address for correspondence
School of Education, University of Sheffi eld, 388 Glossop Road, Sheffi eld S10 2JA. E-mail firstname.lastname@example.org…