Research Of A Teaching Quality Evaluation Method Based On Multivariate Statistical Analysis

In order to reflect the teaching situation accurately,many universities introduce the educational assessment mechanism. Teaching quality assessment is an important aspect of the educational assessment. The teacher who is directly responsible for the teaching quality, plays a dominant role in the implementation of teaching activities. Based on this, the teaching quality assessment is the important content of the Higher Education Quality Evaluation. Teaching quality assessment is a complex issue, involving more content,.Any evaluation method can not be truly comprehensive and objective. How to evaluate the teaching activities reasonably is the key of the teaching evaluation.Factor analysis and principal component analysis are commonly used to evaluate the teaching qualities.However, they are all objective and comprehensive evaluation methods.In order to combine the advantages of objective evaluation and subjective evaluation,this thesis first combines expert assessment method and factor analysis to evaluate teaching quality.Teachers can find their advantages and disadvantages according the factor scores and improve the quality of teaching.In this method,we can also get the comprehensive score based on factor analysis and expert evaluation. In order to achieve the purpose of increase the discrimination of the scores of teachers,this thesis applys the Box-Cox transformation in teaching evaluation.Based on principal component analysis, Expert assessment and principal component analysis are combined for teaching evaluation. Maybe the first principal component can not be used to evaluate the quality of teaching because of the contribution rate of the first principal component is not high enough.But the traditional evaluation functions based on principal component are not reasonable,so we first determined the number of principal component according to the variance contribution rate,then we do the cluster analysis based on the scores of the principal component.In each category, I attempt to rank the teachers based on the teachers’score of the first principal component. Both of the two evaluation methods which combine the advantages of subjective evaluation and objective evaluation make the teaching evaluation more reasonable.