Previously we have discussed models for strong compact and extended radio sources. In this paper we describe how the nature of these models is expected to depend on redshift and make a preliminary attempt to connect theory with observations. In particular we predict that the number of strong compact radio sources should decline more rapidly with redshift than the number of quasars. We also give physical explanations for the steepening of the average spectral index of radio sources with redshift as well as their smaller dimensions. ©1999 American Institute of Physics.