Censuses are widely used to monitor populations. One of the most interesting modern techniques is distance sampling, which depends on some crucial assumptions, including the random distribution of animals with respect to transect lines, which are often asphalt and gravel roads. We tested the assumption of a random distribution of animals with respect to roads for females of ring-necked pheasants (Phasianus colchicus). Roads can be used to census pheasants using distance sampling, without significant bias because pheasants are randomly distributed with respect to roads. Our method is easy to apply to other field conditions where radio tracking data are available and can be used to test the assumption in other studies and/or with different species.