Yulia Tyumeneva - Candidate of Sciences, Senior Research Fellow at The International Laboratory for Education Policy Analysis at the HSE Graduate School of Education, HSE. E-mail: jtumeneva@hse.ru

Alena Valdman - intern-researcher at The International Laboratory for Educational Policy Analysis at the HSE Graduate School of Education, HSE. E-mail: alena.valdman@gmail.com

Martin Carnoy - PhD, Professor at School of Education of Stanford University. Address: Stanford University, School of Education, Stanford, California 94305. E-mail: carnoy@stanford.edu

The purpose of this research is to estimate of relationship between the level of subject knowledge formation with skills of transferring to non-academic context. To answer this question we use Russian databases of international studies TIMSS and PISA. The sample included 4241 Russian students. They participated in both studies TIMSS-2011 and PISA-2012. Based on the previous literature and content comparison of TIMSS-PISA math items we argue that TIMSS items can represent formal subject knowledge, while PISA items represent real-life context problems. We divided TIMSS participants into 6 groups from the top performers to poor performers (based on class average PVs in math) and picked up 10 (and 20) hardest PISA items using One-Parameter Rasch Model. Then we checked the proportion of correctly done PISA items among the 10 (and 20) hardest ones in each TIMSS group of students. This indicator is measure of the ability to transfer subject knowledge in a new context and apply them to solve real-life problems. Positive relationship between the level of subject knowledge development and the ability to transfer and apply it to a every-day problems was found. The better student knows math, the more likely he will be able to apply this knowledge to solve problems in a non-academic context. This relationship is not linear. Students from the most successful TIMSS group had substantive benefit in handling PISA hardest items, but all other levels of TIMSS performance differentiated weakly success in contextual PISA problems.