We analyse the determinants of overeducation among school leavers within a search theoretical framework. Overeducation is measured on the basis of job analysis (JA), the required level to get the job (RL) and realised matches (RM). It is shown that the choice of the measure is crucial for the outcome of the analysis. Job levels measured by RL and RM are clearly biased. Hence, only a few results are robust: higher qualified individuals occupy higher complexity jobs and overeducation is less prevalent among school leavers with better study results and in small organizations. Conversely, JA generally delivers results that are consistent with theoretical considerations. Based on JA, overeducation is also lower among men, whites, school leavers with higher educated parents, search intensive individuals and school leavers from higher quality institutions. Furthermore, overeducated workers are more often hired in occupations without structural bottleneck vacancies and in the financial and professional services sector. Additionally, irrespective of the used measure, heterogeneous requirements within complexity levels explain a significant but small part of the extent of overeducation.
