A Bayesian hierarchical model for comparative evaluation of teaching quality indicators in higher education

Abstract

The problem motivating the paper is the quantification of students' preferences regarding teaching/coursework quality, under certain numerical restrictions, in order to build a model for identifying, assessing and monitoring the major components of the overall teaching quality. We propose a Bayesian hierarchical beta regression model, with a Dirichlet prior on the model coefficients. The coefficients of the model can then be interpreted as weights and thus they measure the relative importance that students give to the different attributes. This approach not only allows for the incorporation of informative prior when it is available but also provides user-friendly interfaces and direct probability interpretations for all quantities. Furthermore, it is a natural way to implement the usual constraints for the model coefficients. This model is applied to data collected in 2009 and 2013 from undergraduate students in the Panteion University, Athens, Greece and besides the construction of an instrument for the assessment and monitoring of teaching quality, it gave some input for a preliminary discussion on the association of the differences in students' preferences between the two time-periods with the current Greek socioeconomic transformation. Results from the proposed approach are compared with the ones obtained by two alternative statistical techniques.

A Heterogeneous Information-Based Multi-Attribute Decision Making Framework for Teaching Model Evaluation in Economic Statistics

“…Regular evaluation of the teaching model should be carried out to ensure continuous improvement of teaching quality [51]. The administration of universities can set up an evaluation group composed of senior teachers to evaluate different teaching models on a regular basis and force the improvement of teaching models according to the evaluation results.…”

Entropy Value-Based Pursuit Projection Cluster for the Teaching Quality Evaluation with Interval Number

“…Fouskakis proposed a Bayesian hierarchical beta regression model with a Dirichlet prior on the model coefficients. The coefficients of the model can then be interpreted as weights and thus they measure the relative importance that students give to the different attributes [ 16 ]. Gu et al analyzed the importance of teaching valuation and the phenomenon of distortion of teaching evaluation data, and proposed a cluster evaluation model for research on teaching evaluation data by using an improved K-Mode clustering method [ 17 ].…”

“…The weighting process is critical for the results of the teaching quality evaluation and it should be guaranteed that the outcomes of the evaluation are not manipulated by those carrying out the assessment [ 16 ]. At the same time, most of the interval number method is complicated and tedious, and sometimes the loss of information in dealing with the problem leads to the distortion of evaluation.…”

A Research on the Quality Evaluation of the Innovation and Entrepreneurship Education at Chinese Universities Based on Linguistic Operators

“…Maria et al, identified the quality determinants for education services provided by higher education institutions (HEIs) in Greece and measured their relative importance from the students' points of view by the multi-criteria decision-making methodology and the analytical hierarchical process (AHP) (Tsinidou, Gerogiannis, & Fitsilis, 2010). Fouskakis, Petrakos & Vavouras, (2016) propose a Bayesian hierarchical beta regression model to build a model for identifying, assessing and monitoring the major components of the overall teaching quality. The above methods cannot be directly used to evaluate the quality of the innovation and entrepreneurship education at colleges and universities, since they have either a weak scientific index system or a simple calculation, or these approaches are not empirically tested.…”