The article is concerned with a process of adjustment between job seekers and vacancies on Poland’s labor market from 1998 to 2010. The process is known as labor market matching. The aim of the paper is to estimate matching function elasticities on the basis of quarterly data with respect to stock variables (unemployment stock and vacancy stock) and flow variables (unemployment inflow and vacancy inflow). The empirical analysis is based on estimates of both random and non-random matching function models, including temporal data aggregation, which makes it possible to determine the main determinants of job creation. The results confirm the importance of flow variables, while not disproving the assumptions of the random matching model, the author says. The latter model may apply to those matches where luck is a key factor in gathering information about job offers, Gałecka-Burdziak adds. The parameter estimates made by the author suggest constant returns to scale. The matching function elasticities in random matching are around 0.6-0.7 with respect to the unemployment stock and 0.3 for the vacancy stock. In the case of non-random matching elasticities, the figures are around 0.6 and 0.1 for the stock variables – unemployment and vacancy respectively – and 0.25 and 0.15 for the inflow of the unemployed and vacancies respectively.