Despite transition started ten years ago, the questions “Who are those unemployed?” and “Does education prevent long-term unemployment?” are still open. The aim of this paper is to determine the major micro and macro factors, which influence the probability of finding a job. The influence of the field of education on the duration of unemployment spell is studied as well. The Russian Longitudinal Monitoring Survey (RLMS) (rounds 5-9) is the basic source for our research. Using duration-data analysis, we estimate proportional hazard models with three groups of factors: demographic (age, gender, marital status), educational (level of education, field of education) and regional (unemployment rate, average wage, metropolitan area). The results of the regressions show that all the three groups of factors significantly influence the probability of leaving unemployment. The hazard rate estimators are compatible with human capital theory. People with higher education tend to exit unemployment faster than their less educated competitors. It is worth noting that people who entered educational institute after 1985 have higher probability to leave unemployment pool. Estimations reveal that those educated in engineering have better chances to get a job. Gender asymmetries in the influence of marital status, small children, and education on the hazard rates are found