Assessment of Performance Unit Science Reports for Teachers

The Assessment of Performance Unit (APU) was set up in 1975 within the Department of Education and Science (DES) to promote the development of methods of assessing and monitoring the achievement of students at school, and to seek to identify the incidence of under achievement. The APU commissioned research teams to create instruments and methods of assessment and to conduct surveys in five curriculum areas: language, science, mathematics, foreign languages and design and technology. Between 1978 and 1988 surveys were carried out in these areas using samples of children aged 11, 13 (science and foreign language only), in England, Wales and Northern Ireland.

Aims and objectives
The APU existed to provide objective information about national standards of children’s performance at all levels of ability. Its terms of reference included making the findings available to those concerned with resource allocations within government departments, LEAs and schools. Hence, as well as formal reports of each survey and academic papers, its outcomes included a series of short booklets for teachers. These provide teachers with brief accounts of what was tested in the APU surveys in science and summaries of performance in different aspects of science of students at different ages and in different sub-groups.

Approach
Certain conditions were established for the surveys. At each age a sample of 1.5 per cent of the students in England (with slightly greater samples in Wales and Northern Ireland) were to be involved in each survey. They were to be selected by a two-stage process: first a selection of schools and then a sample of students within schools. The duration of testing for each student was to be about 45 minutes at age 11 and an hour for the older students.

In view of the limited time for testing, the APU project used a multiple-matrix sampling scheme. A large number of test items were divided into a series of test booklets, distributed across the sample so that each was attempted by a representative sample of students. In science a large bank of test items was created for each category and sub-category being testing and reported. Items for each survey were then randomly selected from each bank so that it was representative of the domain being tested. In addition, information about the context of students’ learning in science was provided by a teachers’ questionnaire which collected information about schools’ environment and resources for science. In surveys at age 11, the questionnaire also asked about the school’s policy and provision for science. At age 15 indications of students’ general ability were collected through enquiry into the number and nature of external examinations the students would take later in the year.