We present the statistical evaluation of a count rate and area limited complete sample of the ROSAT All-Sky Survey (RASS) comprising 674 sources. The RASS sources are located in six study areas outside the galactic plane (|b^II| >= 20 deg) and north of delta = -9 deg. The total sample contains 274 (40.7%) stars, 26 (3.9%) galaxies, 284 (42.1%) AGN, 78 (11.6%) clusters of galaxies and 12 (1.8%) unidentified sources. These percentages vary considerably between the individual study areas due to different mean hydrogen column densities N_H . For the accuracy of the RASS positions, i.e., the distance between optical source and X-ray position, a 90% error circle of about 30 arcsec has been found. Hardness and X-ray-to-optical flux ratios show in part systematic differences between the different object classes. However, since the corresponding parameter spaces overlap significantly for the different classes, an unambiguous identification on the basis of the X-ray data alone is not possible. The majority of the AGN in our sample are found at rather low redshifts with the median value of the whole sample being z_med=0.24. The median values for the individual fields vary between z_med=0.15 and z_med}=0.36. The slope of the log N-log S distributions for our AGN (-1.44 +- 0.09) is in good agreement with the euclidian slope of -1.5. The BL Lacs (with a slope of -0.72+-0.13) exhibit a much flatter distribution.