Operational modal analysis (OMA) is widely used whenever the dynamic characteristics of structuresthat do not fit into a laboratory are needed. In addition, OMA offers a test of the structure under itsreal boundary conditions which can be preferable for validation of numerical models. Theoretically, atleast, the natural frequencies and damping ratios should be identically estimated by an OMA test andan experimental modal analysis (EMA) test. However it is still commonly believed that an EMA testis more reliable because of the available information and controlled environment. The present paperpresents an experimental study of EMA and OMA tests of a Plexiglas plate under free-free conditions.The plate which is a simple geometry is designed to have closely spaced modes and is seen as a goodstructure for comparison. Modal parameters are compared between the EMA and OMA test and it isfound that natural frequencies and mode shapes are similar within 0.2 % (MAC), whereas the dampingratios showed differences up to 7.7 % and are consistently lower for the OMA test. This differences indamping were possibly caused by slightly nonlinear boundary conditions combined with the fact that thevibration levels for the EMA test were higher than for the OMA test.