The Design For Assessment Of Students' Test Grades

It is very important to assess the students'grades of higher normal university, since the students'grades of higher normal university consist of study marks and teaching practice marks, the importance of evaluating students'grades is significant, so assessing the marks of teaching practice should also be valued, teaching practice is an necessary part of higher normal educating, and assessing the marks of teaching practice is also an integral step of teaching practice process. With the further development of educational reform, the traditional model of assessing teaching practice marks exposes its defects in thought and practice, so it's the concerning for teaching practice organizers and educational researchers on how to have a reform and play its expecting role in students'grades assessment.As a student of normal university—a future teacher, his basic skills and comprehensive qualities should be developed to a great extent. This thesis first illustrates the significance of evaluating students'grades, then through reviewing the relevant documents, and raises the problems that the current design for students'grades assessment is facing.The thesis introduces the definition, calculating method and its major functions of the correlation analysis, also analyzes in detail the design for assessing the students'grades based on the relevant analysis method, and makes a thorough examination on the study marks of 14 courses in the first and second semesters about 134 students, (Grade 2002, Mathematics Department, Hulan College, Harbin Normal University)(the original data is provided by the Office of Academic Affairs) in order to find out the Principal component regression and concerning courses and also discusses the rationality of curriculum design.The thesis also illustrates the basic thought, definition, calculating method, nature of Principal component regression and statistic features of main components. The statistics of main components is an usual improvement method for the estimate of linear model parameter, and it can solve the problems brought about by the design of matrix multicollinearity. Assessing the marks of field teaching practice is different from the aptitude test, it reflects a kind of relative index of the whole students, and this demands that it classifies the degree types under the principles of minimal divergence in same category, while maximal divergence in different categories. According to the existing problems and thorough analyzing to them, the thesis puts forward a solving method of comprehensive statistics that is applying the main components analysis to evaluate the marks of field teaching practice.