The Relationship between Economic Insecurity and College Expectations and College Matriculation: Evidence from Russia.

• In Event: Highlighted session: Changes, corruption, and academic integrity in Eurasian higher education

Proposal

How do students respond to economic insecurity in terms of their college expectations and college attendance? A large literature argues that when faced with higher unemployment, students are more likely to attend college because immediately entering the job market is less attractive (Alessandrini, 2018; Barr & Turner, 2013; Clark, 2011). At the same time, higher levels of unemployment result in reduced family income, which may hinder parental ability to pay for college expenses, thereby negatively impacting student propensity to enroll (Conley, 2001). Further, reduced family income may influence whether students are willing to enroll in fee-based education in countries where both tuition-free and tuition-based tertiary education is available.

An important predecessor to students' matriculation into college is whether they expect to achieve a college degree while in secondary school. Taylor & Rampino (2014) show that when unemployment is high, students - especially those from higher-SES families - increase their educational expectations. The authors argue that this is because during economic downturns, expected wages from entry to the job market are low while the incentives to acquire skills to eventually succeed in the job market are high. At the same time, when secondary school students have a pessimistic view of the job market, they tend to disengage in school, which can negatively affect their college plans (Hill, Liang, Bravo, Price, Polk, Perella, & Savitz-Romer, 2018). Further, they might be negatively cued regarding their college prospects from their parents who during economically turbulent times might decrease their expectations towards the level of education that their children will achieve due to a reduced ability to pay for college-related expenses (Behtoui, 2017; Rimkute, Hirvonen, Tolvanen, Aunola, & Nurmi, 2012).

No study has yet assessed how these mechanisms play out in Russia. Russia makes an interesting case for this examination because of the Russian welfare state that might mediate the effect of unemployment, as well as the availability of both free and fee-based higher education (Bain, 2001).

Data: For individual- and school-level variables, we will use Trajectories in Education and Careers Study (TrEC), the first longitudinal study based on a representative sample of high school students in Russia. TrEC is administered by the Centre for Cultural Sociology and Anthropology of the Institute of Education, Higher School of Economics. The initial wave of the study was Trends in International Mathematics and Science Study 2011 (TIMSS 2011), with students being followed up with proprietary surveys four times in subsequent five years. For federal subject-level statistics, we will use the Federal Service for State Statistics (Rosstat), which is a federal unit responsible for collecting official economic and demographic statistical data in Russia.

In this study, we will use three dependent variables: 1) Whether a student expects to attain a university degree; 2) Whether a student is willing to pay for higher education if s/he does not get in to study for free; and 3) Whether a student eventually matriculates into the university. Our independent variable of interest will be unemployment (at the federal subject level). In order to account for potential confounders, we will control for a number of covariates. At the individual level, we will control for student achievement; gender; parental education; parental occupation; and whether a student lives in a two-parent family. At the school level, we will control for whether a student attends an elite school and for school location (rural vs. urban). Elite schools are public schools that are authorized to offer advanced curriculum in a variety of subjects. These schools are also authorized to administer a special entrance examination to those seeking to be enrolled. Further, at the federal subject level, we will control for federal subject GDP. The same independent and control variables will be used across the three models.

Methods: We will use multilevel models with students (i.e. level-one unit) nested within schools (i.e. level-two units), and schools nested within federal subjects (i.e. level-three units). We will use random intercepts at the school and federal subject level. The Russian Federation consists of the following federal subjects: 22 republics, 9 krais, 46 oblasts, 3 federal cities, 1 autonomous oblast, and 4 autonomous okrugs, making it 85 federal subjects. Students from 43 federal subjects are represented in our data, thus ensuring that we have enough higher-level nesting units to allow for robust estimation. We will use survey weights to ensure that our estimates are country-representative. Because our outcomes of interest are binary and follow a Bernoulli distribution, we will use logit models.

Results: Our results will show whether unemployment is associated with a change in (a) student propensity to expect to attain a college degree; (b) in student willingness to pay for higher education studies; and (c) with their eventual matriculation into the university. These results will be discussed with particular attention given to Russian welfare state and the specifics of the Russian higher education system. These results will speak to the literature that investigates the importance of larger economic climate when examining how students make their decisions regarding attending tertiary education. It is important to note that this study will only examine educational expectations and matriculation in terms of university degrees. Future work should investigate if the same patterns would apply to post-secondary technical vocational education.

Authors

• Volha Chykina, University of Michigan

• Katerina Bodovski, Penn State University

• Tatiana Khavenson, National Research University Higher School of Economics