Title: Estimating the number of entities with vacancies using administrative and online data Authors:  Maciej Beresewicz - Poznan University of Economics and Business (Poland)
Herman Cherniaiev - University of Information Technology and Management (Poland)
Robert Pater - University of Information Technology and Management (Poland) [presenting]
Abstract: The number of entities is estimated to have at least one job vacancy. To achieve this goal, an alternative approach is proposed to the methodology exploiting survey data, which is based solely on data from administrative registers and online sources and relies on dual system estimation (DSE). To achieve this, job offers collected from online job boards in Poland and administrative data from public employment services are used. As these sources do not cover the whole reference population and the number of units appearing in all datasets is small, a DSE approach is developed for negatively dependent sources. To achieve the main goal, a thorough data cleaning procedure is conducted in order to remove out-of-scope units, identify entities from the target population, and link them by identifiers to minimize linkage errors. The effectiveness and sensitivity of the proposed estimator are verified in simulation studies. From a practical point of view, the results show that the current vacancy survey in Poland underestimates the number of entities with at least one vacancy by about 10-15\%. The main reasons for this discrepancy are non-sampling errors due to non-response and under-reporting, which is identified by comparing survey data with administrative data.