Bayes Estimation-Structural Equation Model In Education Satisfaction

Satisfaction study is a relatively mature research field, which is used primarily for business, market analysis and management to study consumer satisfaction with a product or service. Satisfaction analysis can apply to any field for providing products and services, including public service. Education as a government public service, in-depth analysis of education satisfaction to help government departments and the public to know exactly the status quo of the current basic education level of satisfaction with the quality of education, help to reveal problems in the development of education lies the contradiction, help around the levels full implementation of quality education, improve the quality of education to provide scientific, rational decision-making reference.Satisfaction research methods are mainly through the establishment of satisfaction index for evaluation. Currently, methods for Satisfaction Index have AHP, fuzzy comprehensive evaluation method, structural equation modeling and so on. These methods mainly solve the weight problem of Satisfaction Index, they have certain advantages for qualitative index, fuzzy index, observed variables and latent variables, but for complex data with a variety of characteristics, especially from the questionnaire data, and the use of a single method may have some limitations. Fusing on the advantage of a variety of methods to handle complex data, not only the results overcome the limitations of a single method, are able to be more objective.The Bayes estimation method not only plays the excellent characteristics in handling small samples, but saves much cost. Structural equation model precisely hands multiple paths to potential variables and observed variables, which are able to analyze detailed and specific reasons for the degree of satisfaction. Based on a questionnaire survey, firstly, using the Bayes estimation method estimates education satisfaction based on the prior distribution of information. Secondly, using structural equation model analyzes the relationship between latent variables and observed variables of education satisfaction. In this paper, a combination of the two methods studies satisfaction, which brings into full play the advantages of Bayesian estimation and structural equation model.This paper takes the questionnaire data of Kunming basic education for example for empirical analysis, the result agrees well with the public consensus. The results prove that, the combination of Bayes estimation model and structural equation modeling utilized for educational quality satisfaction analysis and research is feasible and effective. It broadens the field of satisfaction analysis, and provides a novel idea for multi-index comprehensive evaluation.