Many students in secondary school find physical science and mathematics uninteresting and difficult to learn with understanding. This leaves important gaps in their education and narrows the range of careers open to them. This project will redesign key aspects of the teaching and learning of these subjects, devising a principled approach which is more effective in engaging students and guiding them towards understanding. Insights from several social scientific fields – concerned with conceptual growth, identity formation, classroom dialogue, collaborative learning, and relations between everyday and formal understanding – will guide the design of an intervention suitable for widespread use in normal school settings. This research project will generate tried-and-tested resources for training teachers and teaching students, and improve understanding of teaching and learning processes in science and mathematics. Phase 1 will involve collaboration with teacher co-researchers from several schools to devise and pilot the intervention. In Phase 2, classroom implementation by the teacher co-researchers will be analysed, and the intervention refined accordingly. Phase 3 will evaluate repeated implementation by the teacher co-researchers, alongside initial implementation by teachers from a wider range of schools, compared to the established practice of a control group of teachers from similar schools.

This dataset was collected during the 2010/11 school year as part of a randomised field trial of the epiSTEMe intervention with Year 7 mathematics and science classes in English secondary schools. An open invitation was sent to schools across the Eastern region and into North London. Schools were invited to participate on the basis that teachers from schools later assigned to the intervention group would follow the associated 2-day training programme for the intervention (and receive the associated classroom materials) prior to the field trial; teachers nominated by schools later assigned to the control group would participate in the field trial using their normal teaching approach, before receiving training/materials after its completion. All schools completing the application process were assigned to an experimental group using an approach in which schools were paired according to school type and contextual value-added score, and then randomly allocated between the intervention or control group. One school withdrew prior to the start of the field trial because of staffing shortages. This yielded 25 participating schools; 12 in the intervention group, 13 in the control. Schools were requested to nominate 2 teachers of mathematics and 2 teachers of science, each with a Year 7 class; in the event, not all schools participated in both subjects or nominated 2 teachers in a subject. Schools were also recommended to choose classes in which a majority of pupils had attained level 4 in the relevant subject in end-of-KS2 assessment (a level achieved by around 80% of pupils nationally in mathematics and science). Because many schools did not timetable their Year 7 classes until close to the start of the school year, the assignment of teachers to classes had to take place vicariously within each school without any involvement of the research team. The field trial was scheduled to be undertaken by 70 teachers with a Year 7 mathematics or science class. After attrition of 10 teachers/classes (i.e. insufficient data returns made), the number of teachers/classes included in the analysis was 60: in Mathematics, 12 intervention, 16 control; in Science, 16 intervention, 16 control. In the case of 5 intervention group teachers, data was also collected in a second Year 7 class, but it did not prove necessary to fall back on this data for the analysis. A 25-item attitude questionnaire (in parallel mathematics and science versions) was administered to each participating class at the start and end of the school year. A series of pre-, immediate post- and deferred post-tests tailored to the particular topic were administered to each class when (and if) it studied each of the target topics over the course of the school year. A 20-item opinion questionnaire (in parallel versions for each topic) was also administered to each class after teaching of the target topic was complete. Background data about students was gathered from school records and/or student questionnaire. A table in the documentation shows the 75 classes that set out to participate in the study. Whatever data were collected from these classes are included in the Original data files. As explained above, 60 classes were retained for analysis. For these classes, the variables on which analysis was based are included in the Composite data files. For the 15 classes excluded from the analysis, the reason is shown in the table.