UDC 04.62/37.012.4
DATA SCIENCE TOOLS FOR ANALYZING A LARGE SAMPLE OF STUDENTS TESTING DATA IN MOODLE
A. E. Boyko, post-graduate student, MUCTR, Moscow, Russia;
orcid.org/ 0009-0007-4006-4403, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
T. V. Savitskaya, Dr. Sc. (Tech.), professor, MUCTR, Moscow, Russia;
orcid.org/ 0000-0001-9818-2196, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
D. S. Lopatkin, Ph. D (Econ.), Head of the Department of Management and Marketing, MUCTR, Moscow, Russia;
orcid.org/ 0000-0003-4260-4710, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.
Due to the intensification of processes for building online-courses by national universities using LMS Moodle, where collection, storage, monitoring, and processing of educational data is carried out, the formation of a unified approach to the implementation of descriptive and diagnostic educational analytics that can meet the objectives of quality management system (QMS) in terms of the presenting testing data for the purposes of internal and external quality control of education becomes an urgent task. The aim is to expand the possibilities of monitoring, presenting and visualizing educational data from LMS Moodle related to various forms of testing students using data science tools to optimize educational processes and improve the quality of learning. Results: the paper considers the process of preparation and further processing and visualization of data on diagnostic testing of second-year students in D. Mendeleev University of Chemical Technology of Russia (more than 900 people) in two disciplines. Various options for data aggregation are considered: by groups at the faculty, heterogeneous class, within the educational program, at the level of entire university. Various options for visualizing the results of comparing the academic performance of groups and the tested/total quantity of students are proposed; visualization options and parameters for analyzing distribution "structure" of test scores in study groups at a selected faculty, etc. Practical significance: the results of the study will help teachers and administrators of LMS Moodle to make more informed decisions based on data; the tools proposed can become a starting point for the implementation of basic adaptive approaches in teach-in.
Key words: : : learning analytics, learning methodology system, LMS Moodle, data analysis, Python, Seaborn library, Matplotlib library, descriptive analytics, diagnostic analytics, higher education