Statistical analysis of the aggregated rating of russian universities as a measuring tool

Abstract

In our country, there are several ratings of universities in the field of education, classifying them according to the quality of education. These ratings differ in their goals and approaches to determining what the quality of education is. To combine these approaches, the National Accreditation Agency in the Field of education is developing national aggregated ratings of higher education institutions. The purpose of this work is to provide a statistical analysis of the aggregated ranking of universities in 2021 on the quality of education as a measuring tool. Statistical analysis of the aggregated rating as a measuring tool is carried out within the framework of the theory of latent variables. The aggregated rating is considered as a latent variable, the indicators of which are the eight national ratings. The constructive validity of the aggregated rating is determined. The quality of education in universities was measured on a linear scale. The aggregated rating has more differentiating ability compared to the traditional scoring system. The ratings that differentiate universities with low and high quality of education better than other national ratings are determined. The ratings that are most and least adequate to the measurement model are also shown. A joint analysis of all ratings carried out within the framework of the theory of latent variables made it possible to identify the strengths and weaknesses of each rating. Quality of education in universities is measured on a linear scale; this is important for monitoring and using many methods of statistical analysis involving a linear scale. An important advantage of using the theory of latent variables is the high differentiating ability of universities. It is planned to monitor the universities’ quality of education within the framework of the theory of latent variables.