Research On Evaluation Indexes Based On Generalized Theory And Multifaceted Rasch Model

In recent years,with the continuous deepening of teaching reform,many colleges and universities have used students' evaluation of teaching as an important form and means of teaching quality assessment.Since the authenticity,objectivity and comprehensiveness of data and content are the core of student evaluation,in a sense,students' evaluation of teaching plays a decisive role in teaching management.It is increasingly welcomed by most colleges and universities to put forward high standards for the construction on the index system of students' evaluation of teaching.On the one hand,the target orientation of the index system construction must be clear when constructing the index system,that's say,adopting the development evaluation to improve the ability of the appraisers.To achieve teaching expertise.On the other hand,the entire indicator system should be implemented on the basis of integrity and conciseness.However,in the actual teaching evaluation process,the reliability of evaluation results is not optimistic.This is due to the existence of evaluation indicators beyond the students' evaluation ability or the inaccurate expression of evaluation indicators.Therefore,in order to give full play to the role of mobilizing students in teaching evaluation,we must urgently solve the problem of constructing a set of scientific and reasonable index system of students' evaluation.After initially establishing the index system of students' evaluation,this study randomly selected 15 classes from 16 colleges of Jiangxi University of Finance and Economics,one class each with a total of 32 classes and 1190 undergraduate students.Each class was assigned to five classes through a questionnaire survey.The teacher scores on each evaluation index,and then conducts empirical analysis,revise the evaluation index system,and finally constructs a set of evaluation index system that meets the needs of students.This study is mainly divided into two parts.The first part is an empirical study on the evaluation of generalization theory.The R-software self-compilation program was used to conduct G study and D study,respectively,to explore the sources of teaching errors,as well as the appropriate number of teaching evaluations and the appropriate number of evaluation indicators.The second part is an empirical study on the evaluation of multi-faceted Rasch models.Using the method of random sampling from the evaluation data,10 classes are selected each time and 30 consecutive extractions are performed.Each time a multi-faceted Rasch model is used to estimate the parameters,and the difficulty of each level in each assessment indicator is discussed,each assessment,as well as the fitness of the indicators.The main conclusion of this study are as follows:1.The errors in teaching evaluation come from teachers,teaching evaluation indicators,interaction effects between students,teachers and teaching evaluation indicators,interaction effects between teachers and students,teaching indicators and interaction effects among students,among which the largest source of error is the students' effect.In the actual assessment process,the use of 15 assessment indicators is more appropriate and each class can select 25 students to teach.2.Based on the pretest data of the students,a new evaluation indicator system proposal was proposed.Regardless of the scorer's effect,the evaluation indicator with a fitting index between 0.75 and 1.25 was selected as the recommended teaching evaluation index.Finally,13 indicators of teaching evaluation were selected from 28 indicators of teaching evaluation.3.Based on the analysis results of pre-test data,a scoring evaluation program for teaching options is proposed.In order to distinguish the level of good and bad teachers,consider giving option 4(full consent)the largest weight,and pass the threshold as a weight after linear conversion.