Identification of a complete sample of northern ROSAT All-Sky Survey X-raysources IV. Statistical analysis

We present the statistical evaluation of a count rate and area Limited comp lete sample of the ROS AT All-Sky Survey (RASS) comprising 674 sources. The PASS sources are located in six study areas outside the galactic plane (/b (II)/ greater than or similar to 20 degrees) and north of delta = -9 degree s. 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 source s. These percentages vary considerably between the individual study areas d ue 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 positio n, a 90% error circle of about 30 " has been found. Hardness and X-ray-to-o ptical flux ratios show in part systematic differences between the differen t 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 th e whole sample being z(med) = 0.24. The median values for the individual fi elds 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) ex hibit a much flatter distribution.