We present a detailed description of the first direct measurement of the spatial correlation function of X-ray selected AGN .
This result is based on an X-ray flux-limited sample of 219 AGN discovered in the contiguous 80.7 deg ^ { 2 } region of the ROSAT North Ecliptic Pole ( NEP ) Survey .
Clustering is detected at the 4 \sigma level at comoving scales in the interval r = 5 - 60 ~ { } h ^ { -1 } Mpc .
Fitting the data with a power law of slope \gamma = 1.8 , we find a correlation length of r _ { 0 } = 7.4 ^ { +1.8 } _ { -1.9 } ~ { } h ^ { -1 } Mpc ( \Omega _ { M } = 0.3 , \Omega _ { \Lambda } = 0.7 ) .
The median redshift of the AGN contributing to the signal is z _ { \xi } = 0.22 .
This clustering amplitude implies that X-ray selected AGN are spatially distributed in a manner similar to that of optically selected AGN .
Furthermore , the ROSAT NEP determination establishes the local behavior of AGN clustering , a regime which is poorly sampled in general .
Combined with high-redshift measures from optical studies , the ROSAT NEP results argue that the AGN correlation strength essentially does not evolve with redshift , at least out to z \sim 2.2 .
In the local Universe , X-ray selected AGN appear to be unbiased relative to galaxies and the inferred X-ray bias parameter is near unity , b _ { X } \sim 1 .
Hence X-ray selected AGN closely trace the underlying mass distribution .
The ROSAT NEP AGN catalog , presented here , features complete optical identifications and spectroscopic redshifts .
The median redshift , X-ray flux , and X-ray luminosity are z = 0.41 , f _ { X } = 1.1 \times 10 ^ { -13 } ergs cm ^ { -2 } s ^ { -1 } , and L _ { X } = 9.2 \times 10 ^ { 43 } ~ { } h _ { 70 } ^ { -2 } ergs s ^ { -1 } ( 0.5–2.0 keV ) , respectively .
Unobscured , type 1 AGN are the dominant constituents ( 90 % ) of this soft X-ray selected sample of AGN .