The English branch of the study has been funded by the Government and is being carried out by the National Foundation for Educational Research. It is focusing on only the two younger age groups. This week’s report, which was written by Wendy Keys, Sue Harris and Cres Fernandes, refers to tests that were taken by 3,579 Years 8 and 9 English pupils in March, 1995.

Each pupil took two tests that lasted 90 minutes in total. Just over 80 per cent of the maths questions and three-quarters of the science questions were multiple-choice. Pupils were also asked to complete questionnaires which provide information about their home background and attitudes towards maths and science.

The researchers have also gathered information about the schools that took part in the study and have questioned the children’s teachers about their training and experience, their teaching approaches, and the resources in their classrooms. This contextual information will be discussed in a second report to be published next spring. The nine-year-olds’ test results are expected to be published in the summer of 1997.

The IEA has gone to great pains to ensure that these studies are the most representative comparisons that have ever been produced. Nevertheless, some assessment specialists point out that it is virtually impossible to establish an absolute international standard for good mathematical knowledge at a particular age.

Obtaining wholly representative samples is also extremely difficult. The TIMSS sampling followed rigorous procedures that were vetted by independent referees, and countries that do not meet the required standards are being excluded from the tables.

Even so, there is probably no such thing as a perfect sample. The NFER originally selected 150 schools for the Years 8 and 9 study, along with two matching sets of 150 replacement schools. But only 127 schools actually took part (85 first-choice schools and 42 replacement schools), a response rate of 85 per cent.