Institute of Education

Publications

• Book

  Kirill Ditkovsky, Fridlyanova S., Gokhberg L. et al.

  Science. Technology. Innovation: 2024 : Pocket Data Book

  This pocket data book contains main S&T and innovation indicators for the Russian Federation. The publication includes the most recent statistical data on R&D input and output, as well as international comparisons. The data book includes information of the Russian Federal State Statistics Service (Rosstat), Ministry of Science and Higher Education of the Russian Federation, Russian Federal Service for Intellectual Property (Rospatent), Organisation for Economic Co-operation and Development (OECD), European Statistical Office (Eurostat), UNESCO, World Intellectual Property Organisation (WIPO), national statistical offices of other countries, and results of methodological and analytical studies of the HSE Institute for Statistical Studies and Economics of Knowledge. In some cases, 2022 data are preliminary.

  National Research University Higher School of Economics, 2024.

• Article

  Sorokin P. S., Мироненко И. А., Журавлев А. Л.

  Boris Parygin’ Personality Social Psychology

  International literature on the history of Russian psychology is largely limited to the development of Activity theory and cultural-historical theory. This paper aims to go beyond these limits by introducing a figure little known for an international audience, Boris Parygin, who emerged in the 1960-s with his “grand” project for the development of Russian social psychology, significantly different from the Soviet psychology “mainstream”, stemming from Vygotsky, Luria and Alexei Nikolaevich Leontiev. We demonstrate that Parygin's works belong to a theoretical and methodological tradition of social and humanitarian scholarship that had been developing in Russia since 1870s. This was a broad and inter-disciplinary stream of theoretical and methodological developments, including authors like Berdyaev, Mikhailovsky, etc., that embraced various disciplines (primarily, psychology, sociology and philosophy).

  Despite the fact that Parygin’s project did not receive substantial support from the soviet academic community, his ideas concerning personality and emphasis on the importance of individual choice and self-determined activity mediating the impact of social surroundings, have become relevant in the context of current discussions in the international science about individual agency, including debates concerning the transformative potential of the individual upon the social structure under conditions of “morphogenetic society” (M.Archer) or “neo-structuration” (P. Sorokin).

  Journal of the History of the Behavioral Sciences. 2024.

• Book chapter

  Lysova-Golomzina E., Kapuza A.

  VALIDATION OF THRESHOLD CONCEPTS IN CASE OF SOCIAL SCIENCE AT MIDDLE SCHOOL USING CONCEPT MAPS

  In bk.: 16th International Conference on Education and New Learning Technologies, Palma, Spain. 1-3 July, 2024. EDULEARN24 Proceedings. IATED, 2024. P. 2290-2295.

• Working paper

  Kardanova E., Ivanova A., Tarasova K. et al.

  A Novel Psychometrics-Based Approach to Developing Professional Competency Benchmark for Large Language Models

  The era of large language models (LLM) raises questions not only about how to train models, but also about how to evaluate them. Despite numerous existing benchmarks, insufficient attention is often given to creating assessments that test LLMs in a valid and reliable manner. To address this challenge, we accommodate the Evidence-centered design (ECD) methodology and propose a comprehensive approach to benchmark development based on rigorous psychometric principles. In this paper, we have made the first attempt to illustrate this approach by creating a new benchmark in the field of pedagogy and education, highlighting the limitations of existing benchmark development approach and taking into account the development of LLMs. We conclude that a new approach to benchmarking is required to match the growing complexity of AI applications in the educational context. We construct a novel benchmark guided by the Bloom's taxonomy and rigorously designed by a consortium of education experts trained in test development. Thus the current benchmark provides an academically robust and practical assessment tool tailored for LLMs, rather than human participants. Tested empirically on the GPT model in the Russian language, it evaluates model performance across varied task complexities, revealing critical gaps in current LLM capabilities. Our results indicate that while generative AI tools hold significant promise for education - potentially supporting tasks such as personalized tutoring, real-time feedback, and multilingual learning - their reliability as autonomous teachers' assistants right now remain rather limited, particularly in tasks requiring deeper cognitive engagement.

  Computation and Language (cs.CL); Artificial Intelligence (cs.AI). cs.CL. arXiv, 2024