ИСТИНА

Graduation examination is an important milestone in life of every school student because it determines opportunities for further education. To graduate from school, Russian students have to pass a series of Unified State Exams (USE) that assess the knowledge of specific school subjects. Our previous study explored cognitive and non-cognitive predictors of USE performance (Voronin et al., 2018). That study revealed that learned mathematical skill was the only significant predictor of Mathematical USE performance after distinguishing regular and specialized schools. In the current study we explore cognitive predictors of USE performance on an independent sample. By these terms, we compare Russian USE to National Scholarship Test (ORT) in Kyrgyzstan, patterned after SAT. The sample included 260 students from Russia (mean age 17.4, SD = 0.4 years) and 805 students from Kyrgyzstan (mean age 17.5, SD = 0.4 years). Russian and Kyrgyz students passed similar assessment of mathematical ability (Number Line, Number Sense), learned mathematical skills (Understanding Numbers, Problem Verification Task), working memory (Corsi Block), reaction time and fluid ability (Raven’s Standard Progressive Matrices). Mathematical and Russian USE scores were available for Russian students, Basic and Mathematical ORT scores were available for Kyrgyz students. We revealed substantial variability or USE and ORT scores both within and across classes. Cognitive measures explained 36% and 17% of Math and Russian USE and 21% and 3% of Basic and Math ORT. Fluid ability predicted Math USE and Basic ORT, learned mathematical skills predicted Russian USE and Basic ORT.