Прогнозирование успеваемости в вузе по результатам ЕГЭ
This paper presents the results of econometric study on predicting first-year grade point average and dropout probability with national examination (EGE) results at the bachelor program in Economics in the Higher School of Economics. The use of a sum of four exams - math, social studies, Russian language and foreign language - shows worse fitting than a sum of three exams, excluding foreign language. In models with separate exams as regressors, the greatest effect on dependent variable provides math grades.
The new system of university admissions in Russia has simplified the process of university entry by decreasing transaction costs associated with the university choice and application process. While procedure now allows applicants to apply to several higher education institutions at the same time, many students do not take advantage of this opportunity and apply only to a single university. In this study we analyze the factors that influence application strategies, whether to apply to only one institution or to apply to several. We argue that higher scores at Unified State Examination (USE) predict a higher probability of multiple applications. Additionally, graduating from a high school that offers advanced training in a particular discipline positively influences this probability. The variables of family income and social capital, a parent’s level of education, and their age, as well as attending additional programs of pre-entry training are statistically insignificant. Thus we demonstrate that while the access is widening, some prospective students do not use all opportunities associated with the new mechanisms of access.
The aim of this paper is to present a case study in the analysis of university applications to the Higher School of Economics (U-HSE), Moscow. Our approach uses lattice-based taxonomies of entrants’ decisions about undergraduate programmes. These taxonomies were built by means of Formal Concept Analysis (FCA). FCA is a well-known algebraic technique for objectattribute data analysis. Admission data as well as formalised survey data were used to reveal possibly significant factors of entrants’ decisions. In this paper we argued that institutional characteristics of the admission process are highly correlated with entrants’ choice. The obtained results are helpful to the university to correct the structure and positioning of its undergraduate programmes.
Starting from 2008, the admission to the BSc program at the ICEF (NRU HSE) is done by the UNE results. 2008 was a transition year: both UNE and internal HSE exams results were counted. Since 2009 the regulations stay stable — only the UNE and Olympiads’ results have been taken into account. In the article the academic success factors for the students admitted in 2009–2011, after 1, 2, and 3 years of study, are analyzed. It has been shown that the students — Olympiads’ winners consistently show better results than other students, other factors equal. The UNE results are still significant for academic performance forecast after 3 years of study. But the final rating of the student after the first year accumulates almost completely the information which the UNE and Olympiads results contain. Region of a student’s school is not significant and gender is marginally significant for future academic performance.
The paper considers efficiency of the Russian Unified State Exam (USE, or EGE, Russian version of SAT), and results of «olympiads» - national students' contests in economics, mathematics as predictors of ICEF HSE students' academic achievements. Most courses in ICEF are taught in English, but USE in English is not a significant predictor. USE in mathematics and Russian are significant predictors, but only USE in Russian is significant for the probability of drop-out. Olympiads' winners demonstrate substantially better performance than other students.
The article examines the main trends in the study of the Stalinist period and the phenomenon of Stalinism in connection with the mass opening of the archives.