Big Data in Higher Education

The development of educational analytics is becoming a key technological trend in higher education, according to foreign experts. How are Russian universities implementing these innovations? Examples of using big data in the educational process were presented at the second international conference "Big Data in Education," organized by Moscow City Pedagogical University. The conference demonstrated how learning analytics contributes to improving the quality of education and enhancing the effectiveness of the educational process in Russian universities.

The conference focused on research into digital platforms in schools. We highlighted five papers devoted to the implementation of digital technologies in universities. They demonstrate how the use of big data can significantly improve the efficiency of various processes at the university, from the development of educational programs to the assessment of student projects. These presentations emphasize the importance of digitalization in higher education and its potential for optimizing the educational process.

1. Design educational programs relevant to the labor market

Universities determine the curriculum for future linguists, programmers, and petroleum engineers based on professional standards, Federal State Educational Standards (FSES), and educational programs. However, this chain has long been criticized for its slow response to changes in professional practices. Recent legislative changes have given developers of educational programs at universities and colleges the opportunity to combine competencies from various fields. This innovation allows graduates to acquire multiple qualifications, making them more competitive in the labor market. As a result, educational institutions are becoming more adaptive to modern requirements, which has a positive effect on the quality of specialist training.

Study additional materials:

The practice of forming Federal State Educational Standards (FSES) based on professional standards will be changed. These changes are aimed at improving the quality of education and meeting labor market requirements. Updated approaches to developing the Federal State Educational Standard (FSES) will provide greater flexibility in educational programs, better preparing students for professional work. As a result, graduates will possess the necessary competencies and skills that meet modern requirements. Changes in the practice of developing the FSES will open up new opportunities for educational institutions and ensure a higher level of specialist training.

To obtain information about necessary competencies, in addition to professional standards, it is worth paying attention to the opinions of employers. However, it is important to keep in mind that the opinion of an individual employer does not always reflect general trends in the labor market. Mikhail Sverdlov, Development Director at Skypro, suggests creating skill sets for various professions even before the development of an educational program. This will help more accurately identify the relevant skills and knowledge necessary for a successful career in a particular field. Using skill sets will allow educational institutions to better adapt their courses to market demands and prepare specialists who meet modern realities.

A skill set is a set of in-demand competencies necessary for successful professional work. Skill sets are developed based on an analysis of labor market needs, which emphasizes the importance of adapting educational programs to current requirements. According to Sverdlov, educational content should focus specifically on these competencies in order to prepare specialists capable of effectively competing and developing in their field.

I believe that the only correct approach to developing educational programs is their adaptation to current market requirements. Otherwise, the question arises: why are we training students if the course content does not meet current or projected needs? It is important to understand what skills and knowledge are in demand in the labor market today and will be relevant in the next four years, when our graduates begin to build their careers.

Big data collection and analysis methods play a key role in developing skill sets. In Sverdlov's project, dedicated to creating a data analyst training program for Skypro, the data volume was insignificant, but this approach can be effectively applied at various scales. The process includes three key steps.

Data collection is the basis for further analysis. At this stage, it is important to identify information sources and collect data that will be useful for developing analyst skills. Next comes the analysis phase, during which the data is processed and interpreted. This helps identify key trends and patterns that will aid in the development of effective learning materials. The final step is to apply the findings to create a structured curriculum that meets current market demands and student needs.

Using such methods not only improves the quality of instruction but also promotes the development of professional skills necessary for data analysis.

a survey of specialists hiring for a vacancy that matches the future educational program;
analysis of job postings on career portals—how many are posted, how quickly they are filled, what requirements they impose;
a detailed manual analysis of several vacancies.

Defining the key knowledge and skills required by a specialist in the modern labor market allows for more accurate development of educational programs. Understanding which competencies are essential and which are secondary allows for the effective design of course content and scope. This contributes to the creation of relevant and in-demand educational programs that will help students successfully adapt and compete in the professional environment.

Reworked text:

Study additional materials:

The ADDIE model for designing educational programs in higher education is a structured approach that includes five key stages: analysis, design, Development, implementation, and evaluation. This model helps create effective courses and programs that meet the needs of students and the demands of the modern labor market.

At the analysis stage, it is important to determine the learning objectives, student needs, and the context in which the learning will take place. Design includes developing the course structure, selecting teaching methods and tools, and forming assessment criteria. During the development stage, learning materials are created, including lectures, assignments, and additional resources.

Implementation involves implementing the course, training instructors, and providing the necessary support to students. Finally, the evaluation stage allows you to analyze learning outcomes, identify the strengths and weaknesses of the program, and make adjustments to improve its effectiveness.

Using the ADDIE model in the design of educational programs helps create high-quality and targeted learning that meets the needs of students and educational institutions.

2. Admitting Motivated First-Year Students

To successfully implement big data analysis in university admissions, data must be collected both at the educational institutions themselves and during the school-age stage. As Alexey Semenov, Academician of the Russian Academy of Sciences and Russian Academy of Education, noted at the conference, universities can effectively utilize the "digital footprint" of applicants. This approach will allow for a more accurate assessment of candidates' potential and their preparedness, which in turn will improve the quality of student recruitment. The use of big data will help optimize the admissions process, making it more transparent and adaptive to the current requirements of the educational system.

Analyzing students' academic performance allows us to predict their success at university. This can become an alternative to the Unified State Exam. By studying how a student mastered mathematics in middle and high school, we can confidently predict how they will cope with their studies at a chosen university. This is based on data on other students who have applied to this university over the past few years with different backgrounds. In this way, we can offer recommendations for continuing education without having to rely on strict selection criteria.

Photo: Juice Flair / Shutterstock

To implement this idea, schools need digital platforms that can reliably track the educational progress of each student. Such platforms should record learning goals, assignment completion both digitally and in the classroom, as well as the time and quality of independent work. It is important to consider how students interact with peers and teachers, as well as receive feedback to analyze and improve the learning process. Furthermore, similar analytics systems should be developed for higher education institutions, allowing for the prediction of student academic performance based on their initial data. This will help educational institutions more effectively adapt curricula and teaching approaches based on analytical data.

Technical solutions for collecting data in the educational process already exist. However, the widespread implementation of these technologies in Russian schools and the use of big data for university admissions are only 10-15 years away. Digital portfolios of student achievements, which Rosobrnadzor plans to implement in the coming years, will merely complement the existing university admissions system. These portfolios will allow for the recording of student success in Olympiads, competitions, and volunteer activities. At the same time, academic performance will continue to be assessed using the Unified State Exam. It is important to note that the integration of digital tools into the educational process can significantly improve the transparency and objectivity of student assessment.

3. Predict and Prevent Dropouts

Russian universities are actively implementing big data analytics related to the educational process. Educational institutions are not only adapting existing commercial tools for these purposes but also developing their own technical solutions. At a conference, Roman Kupriyanov, Deputy Head of the Information Technology Department at Moscow State Pedagogical University, shared his successful experience in this area. The implementation of big data analytics allows universities to more effectively manage education, improve the quality of education, and make informed decisions based on the collected information.

Researchers at Moscow State Pedagogical Univ. have developed an innovative system capable of predicting student academic performance in the next semester with 71% accuracy. The model analyzes various data, including Unified State Exam (USE) scores, which reflect students' initial knowledge, as well as the results of previous sessions. Student participation in social activities and the use of library resources, both physical and electronic, are also important factors. Kupriyanov noted that this system is already demonstrating its effectiveness and is bringing tangible benefits to both students and faculty.

The use of modern student performance monitoring systems contributes to a significant reduction in the expulsion rate. At our university, we have achieved successful results, cutting the percentage of expulsions for academic failure in half. Regular academic performance monitoring allows us to identify problem areas and provide timely assistance to students, which, in turn, increases their motivation and facilitates successful completion of their studies.

MSPU continues to develop its system, including integration with a digital recommendation platform. Currently, the academic performance forecast is available only to university staff. In the future, the system will be directly linked to students' personal accounts, allowing them to track their progress and develop individual plans to address any issues that arise. This will improve students' engagement with the educational process and help them achieve their academic goals more effectively.

4. Manage class attendance

Universities such as MSPU typically have all the student data by default. At the conference, Marina Lapenok from Ural State Pedagogical University presented an attendance prediction model based on the analysis of personal information and individual behavioral characteristics of students. This model can help educational institutions more effectively track student attendance and engagement, which, in turn, contributes to improving the quality of education and increasing academic performance.

To determine the likelihood of students attending classes in a particular subject, relying solely on their previous grades is not enough. The USPU system takes into account various factors, such as a student's temperament, employment status, and whether their interests match the chosen subject. This data was collected through a survey. Research conducted at the university showed that the most accurate predictors of attendance are motivation to study and self-organization. This finding opens up new possibilities for further development of the system, noted Lapenok.

Training the neural network opens up new possibilities, such as generating recommendations for improving student learning and optimizing the schedule, which contributes to increased class attendance. By continuing to develop neural networks, we will be able to more effectively adapt educational processes to the individual needs of students, which will ultimately lead to improved academic results and student engagement.

Photo: BoxerX / Shutterstock

The development of USPU is currently a graduation project of one of female undergraduate students. The model was created and tested using data collected during her training. In the future, the university may consider introducing similar measurements on a larger scale, which will improve the educational process and increase its effectiveness.

Study additional materials:

At the Russian conference on information and educational technologies for universities, current issues of introducing modern technologies into the educational process were discussed. Conference participants shared experiences using digital tools, such as online courses and distance learning platforms. The need for advanced training for teachers in information technology was also emphasized. EdTech development trends and the impact of technology on the quality of education were discussed. Experts shared successful cases of integrating innovative solutions into the educational process and examined the challenges and prospects facing higher education institutions in the context of digitalization.

5. Assess what's not reflected on paper

The growth of data volumes leads to an increase in the number of questions related to legal aspects that fall into the "gray zone" of legislation. This was stated by Ruslan Suleimanov, Head of the Information Technology Department at Moscow City Pedagogical University, in his report on video analytics. In the context of rapid technological development, it is important to consider the legal nuances and potential risks associated with the processing and use of big data.

Moscow City Pedagogical University (MCPU) is developing unique algorithms for assessing student engagement based on video recordings. The university's current solutions, which utilize some open-source developments, allow for the analysis of the behavior of one to four participants in a video. The algorithms record parameters such as posture, gaze direction, and facial expression. These technologies can be applied to both online and in-person lessons recorded in 360° format. The development by Moscow State Pedagogical University opens new opportunities for improving the quality of the educational process and monitoring student engagement.

In the future, video analysis systems will have a significant impact on the educational process, allowing instructors to more effectively evaluate group project work. They will help identify the activity of each team member, which is often difficult to achieve without additional tools. Video analysis will provide valuable information about student interactions and their contribution to the overall work. This will allow for more accurate assessment of not only final results but also learning processes, which are not always visible in final projects. Thus, the use of such technologies can significantly improve the quality of the educational process and ensure a fairer assessment of students' work.

Big Data in Higher Education

Contents:

1. Design educational programs relevant to the labor market

2. Admitting Motivated First-Year Students

3. Predict and Prevent Dropouts

4. Manage class attendance

5. Assess what's not reflected on paper