¹ See, for example, Byrne, E. M., Women and Education (London, 1978);Google ScholarDeem, Rosemary, Women and Schooling (London, 1978);Google ScholarEntwistle, N. and Wilson, J., Degrees of Excellence: the academic achievement game (London, 1977);Google ScholarMaccoby, Eleanor, The Development of Sex Differences (Stanford, 1966);Google ScholarSharpe, Sue, Just Like a Girl (Harmondsworth, 1976);Google Scholar and Sutherland, M. B., Sex Bias in Education (Oxford, 1981)Google Scholar.

² There are exceptions. For example, in the Kweneng District of Botswana, two girls attend primary school for every boy, reflecting the system of male adult labour migration which forces many young boys to become permanent herdsmen. See Allison, Caroline, ‘The Economics of Household Demand for Children's Schooling: study of the Kweneng District of Botswana’, in Development Research Digest (Brighton), 4, Winter 1980, pp. 55–7Google Scholar.

¹ Ministry of Education, Yearbook (Yaoundé, 1970)Google Scholar.

² Santerre, Renault, Pédagogie musulmane d'Afrisque noire (Quebec, 1974)Google Scholar.

³ See, for example, Hake, J. M., Parental Altitudes toward Primary Education in a Hausa Community in Northern Nigeria (Kano, 1970);Google ScholarBlakemore, Kenneth, ‘Resistance to Formal Education in Ghana: its implications for the status of school leavers’, in Comparative Education Review (Madison), 19, 2, 1975, pp. 237–51;Google Scholar and Dubbeldam, L. F. B., The Primary School and the Community in Mwanza District, Tanzania (Groningen, 1970)Google Scholar.

⁴ Mbilinyi, Marjorie, ‘Education, Stratification and Sexism in Tanzania’, in The African Review (Dar es Salaam), 3, 2, 1973, pp. 327–40Google Scholar.

⁵ Mateossian, B., La Population du pays Bemiélké et des départements limitrophes: principaux résultats de l'enquête démographique de 1965 (Paris, 1966)Google Scholar.

¹ Labrousse, André, Le Financement de l'enseignement du premier degré au Cameroun oriental (Paris, 1975)Google Scholar.

² Ministry of Education, Yearbook (Yaoundé, 1975)Google Scholar.

³ Mbassi-Manga, François, ‘Rapport d'activités’, Faculté de Lettres et de Sciences Humaines, Université de Yaoundé, 1974Google Scholar.

⁴ Direction de la statistique et de la comptabilité nationale, L'Enseignement au nord Cameroun (Garoua, 1972). In all Provinces girls have lower pass rates than boys, but the difference between the sexes is smallest in the two Provinces with the lowest enrolment levels, the East and the North. In the North so few girls take the C.E.P.E. that we would expect those who do to be either extremely intelligent or from unusually high social backgrounds, or a combination of both. Remi Clignet and Philip Foster have shown that girls attending secondary schools in the Ivory Coast are, on average, from higher socio-economic backgrounds than boys; The Fortunate Few (Evanston, 1966). For Nigerian data, see Shoremi, M. and Mott, F., ‘Characteristics of Expectations of Lagos University Undergraduates’, Human Research Unit, University of Lagos, 1972Google Scholar.

¹ Ministry of Education, Yearbook, 1970 (Yaoundé, 1970)Google Scholar.

² S.E.D.E.S., Enquête sur le niveau de vie à Yaoundé (Paris, 1974), p. 83Google Scholar.

³ Sample data show that only small traders send substantially more girls than boys to primary school.

⁴ See Cooksey, Brian, ‘Education and Class Formation in Cameroun’, Ph.D. dissertation, Centre of West African Studies, University of Birmingham, 1978Google Scholar.

¹ C.E.P.E. pass rates gave the boys only an 8 per cent advantage over the girls.

² From our discussion it is clear that marking conventions exaggerate real performance differences between the sexes. Here we are interested in making comparisons across occupational groups, holding the built-in bias constant.

³ For a more detailed analysis of performance variations, see Cooksey, Brian, ‘Social Class and Academic Performance: a Cameroon case study’, in Comparative Educational Review (Chicago), 25, 3, 10 1981, pp. 403–18Google Scholar, and Blakemore, Kenneth and Cooksey, Brian, A Sociology of Education for Africa (London, 1980), pp. 64–5Google Scholar.

¹ Between 14 and 35 per cent of candidates from these three groups live in good material conditions, compared with zero and 8 per cent of unskilled and semi-skilled workers' children, and 62 and 84 per cent of those from large-trader and élite backgrounds. We found that 72 per cent of unskilled manual workers had no educational diploma compared with none of the élite.

¹ Foster, Philip, Education and Social Change in Ghana (London, 1965)Google Scholar, and Beckett, Paul and O'Connell, James, Education and Power in Nigeria (London, 1977)Google Scholar, among others stress the importance of parental education in determining patterns of selectivity, but do not stress the sexual dimension.

¹ For example, Sharpe, op.cit.

² The pass rates for boys aged 10–12 and 13–14 were 28 and 24 per cent, respectively, and for girls 25 and 9 per cent.

³ Sharpe, op.cit. p. 145.

⁴ Ibid. p. 135.

¹ Vagliani, W. Weekes, Family Life and Structure in Southern Cameroon (Paris, 1976), p. 32Google Scholar.

² Over 40 per cent of new entrants into this kind of schooling were from outside Yaoundé, but only 15 per cent of those recruited from outside were girls.

³ Girls account for only one-tenth of the final-year students in private academic secondary schools. See Ministry of Education, Yearbook, 1975, p. 45.

¹ If the quality of primary schooling and teaching affect performance levels, then this finding is not surprising: élite children may begin to under-perform at the secondary level, and the effect of this will be more noticeable among girls since those in this category form a higher proportion of the total number of students than boys.

² Cooksey, ‘Social Class and Academic Performance’.

³ Lloyd, Barbara B., ‘Education and Family Life in the Development of Class Identification Among the Yoruba’, in Lloyd, P. C. (ed.), The New Elites of Tropical Africa (Oxford, 1966), pp. 263–83Google Scholar.