The paper describes the method of “compression” of information by allocating latent variables from Big Data with the results of surveys. The peculiarity of the data used to compress information is the presence of various types of answers in an arbitrary textual form, which does not allow the use of existing techniques in which the evaluation of respondents’ answers is carried out only on the Likert scale. Based on the described technique, it is possible to convert any database with survey results into a form suitable for statistical processing, and to provide compression of the initial data array to sizes that allow a comparative analysis of survey results over a long period of time. Analysis of the latent variables obtained will allow the university management to assess the demands of future engineering teachers, constantly monitor trends in changing requirements, identify weaknesses in the organization of university activities, as well as design and implement appropriate corrective and preventive actions in a timely manner. The paper represents the results of testing the technique on the example of building the Student Satisfaction Index. The results of the data analysis have confirmed the validity and reliability of the results obtained.
CITATION STYLE
Kovalenko, O., Bondarenko, T., Kupriyanov, O., & Khotchenko, I. (2020). Allocation of Latent Variables from Big Data in Institutional Researches of Engineering Teachers. In Advances in Intelligent Systems and Computing (Vol. 1135 AISC, pp. 288–296). Springer. https://doi.org/10.1007/978-3-030-40271-6_29
Mendeley helps you to discover research relevant for your work.