Date of Award


Document Type



Edith Cowan University

Degree Name

Doctor of Philosophy


Faculty of Community Services, Education and Social Sciences

First Supervisor

Mark Hackling


The purpose of this research was to develop, implement and evaluate science investigation work based on a cognitive apprenticeship model of instruction and 'linked' to different assessment procedures. Data were gathered to evaluate the effects of the instruction and approaches to assessment on teachers and students, and on the development of students' investigation competency. Year 9 students performed six science investigations in which they examined a relationship between variables. Groups of three students worked together to choose the independent variable, plan their investigation, collect and analyse their data, and to evaluate their findings. The cognitive apprenticeship model of instruction included teacher modelling, coaching, scaffolding and fading, articulating and involving students in self-reflective and metacognitive practices. The three classes which participated in the study (n = 66) experienced different assessment regimes, teacher assessed and norm referenced, teacher assessed and criterion referenced and student assessed and criterion referenced assessment. The study was a naturalistic inquiry and data were collected from numerous sources including a pre and pastiest pencil and paper Test of Science Investigation Skills, pre and post test investigations in which students were assessed from their written responses on an Investigation Planning and Report Sheet, student questionnaires, teacher and student group interviews, and audio and video data. The qualitative data were summarised and interpreted as 76 assertions relating to the themes of the research; investigation competencies, the cognitive apprenticeship model of instruction and the assessment regimes. Sixteen general assertions which were considered to be more genera research findings were then formulated. Students' investigation competencies improved significantly as determined by their pre and post test performances on the Test of Science Investigation Skills, and by their performances on the Investigation Planning and Report Sheet. In addition to improving specific investigation competencies such as planning and conducting investigations, processing data and evaluating investigations, students also perceived that they improved social and workplace skills including working cooperatively, attending to detail, managing time and being organised. Difficulties students experienced in performing investigations were also identified. Based on improved students' performance, the cognitive apprenticeship model of instruction could be deemed as effective in teaching and learning science investigation competencies. As implemented in the study, weaknesses in the implementation of teacher modelling were exposed and it was suggested that modelling would be more beneficial if it were to occur in response to requests from students rather than at the behest of the teacher as an introduction to a learning activity. Classes experiencing teacher assessment, both norm referenced and criterion referenced, achieved similar gains in investigation competencies and the feedback to students following these assessments was similar in quality and quantity. Students in the student assessed criterion referenced class made more modest improvements in Planning investigations and Conducting investigations. These students lacked opportunities for high quality teacher feedback. Clearly these data need to be interpreted with caution because other factors in the learning milieu were not controlled. The implications arising from the research for classroom practice addressed factors contributing to science investigation competency, the cognitive apprenticeship model of instruction and the assessment of science investigations.