Exam board study finds chance of ethnic bias
At present, A-level candidates must write their names at the top of their scripts, although some simply write an initial and the surname. Fears have repeatedly been expressed that awareness of the candidate's sex or ethnic origin could influence the examiner's perception of the paper, even if this were subliminal.
Arguments against introducing blind marking include the administrative cost and the fact that it is often possible to guess the sex of the candidate from the handwriting. The AEB study found this latter point to be true - examiners guessed the sex of the writer from the handwriting style in 70 per cent of cases (girls tend on average to use a more rounded script; boys' writing is freer and more untidy).
The AEB carried out two experiments in blind marking in A-level English literature and A-level chemistry. These subjects were chosen because they are most likely to harbour stereotypes - girls good at English, boys at science. The 24 examiners taking part in the study were each given 30 scripts. The content of each examiner's script was identical to the content of every other participant's script; only the presentation was altered to see if it had any effect on the marks awarded. Each script was therefore available in four forms: male name, male handwriting; male name, female handwriting; female name, male handwriting; female name, female handwriting. The same scripts presented "blind" were given to other examiners as a control. Participants in the experiment were not told its purpose.
The study found that marks were not affected in either English or chemistry by the gender of the name on the script. Nor were they affected by the gender style of the handwriting. Male English literature marks were just as good as female, and female chemistry marks were as good as male.
Jo Anne Baird, who wrote the report, suggests that examiners may be very unlikely to nurture simplistic stereotypes about male and female performance at A-level anyway: "Even if you did think that males were weaker than females in general at English literature, you may understand that those males who are not good at English will not be very well represented in candidates taking English."
The AEB's findings contradict earlier work on the issue, which has frequently found that while women are not judged as favourably when their performance is very good, they attract leniency when their performanc e is poor. In 1984, 11 schools took part in another study in which science teachers were asked to assess a piece of work. Names of pupils were mixed up randomly. Teachers judged the "boys" more highly in richness of ideas, level of interest, suitability for further study, scientific accuracy, organisation and conciseness. The scripts which the teachers believed were written by girls were praised for "neatness".