prejudice in schools. Of course, such pre-assessment affect exam entries: 25% of Affro-Carribean boys were entered for 'O' levels, compared with 53.4% of white boys (Eggleston et al, 1986) - this is despite findings of equal, or superior intelligence: Bagley (1975) found children of middle-class African origin performed better on a Stanford Binet IQ test than similar white children. If a black pupil is female or has special needs, his/her chances of succeeding or even slimmer.It is not only girls of ethnic minorities who are prevented from achieving their full potential - so too are middle class, white girls in 'good' schools. Even today, the dominant, albeit unconscious, ideology is to regulate cultural norms of gender behaviour (Walker and Barton, 1983); these norms are carried into school by both staff and pupils. Girls are not expected to do as well, are not brought up to be assertive, and so tend to underachieve, and blame themselves for it (Light and Dwek, 1987). Teachers praise different qualities in boys and girls - girls are rewarded for being 'obedient, tidy and conscientious' rather than 'lively, independent and energetic' (Arnot, 1981). Until very recently, girls' choices in the curriculum were limited, either explicitly or implicitly, and they were encouraged not to choose scientific or technical subjects: Deem (1978) states that girls' exam successes cluster around arts subjects, which are of limited value for entry into jobs, and lead to different routes in further education. Exams themselves and other measures of achievement are biased against girls: for example, multiple choice tests are much more suited to 'male' logic than 'female' verbal abilities. Problems such as these make it important that steps are taken to change prejudiced expectations and the ways in which pupils are assessed; this would include educating teachers to understand their biases, and helping girls to alter their expectations so that they can achie...
