Random copolymers made up of subunits with arbritary degrees of flexibility are useful as models of biomolecules with different kinds of secondary structural motifs. We show that the mean square end-to-end distance [R-2] of a two-letter A-B random heteropolymer in which the constituent polymeric subunits are represented as continuum wormlike chains and the randomness is described by the two-state Markov process introduced by Fredrickson, Milner, and Leibler [Macromolecules 25, 6341 (1992)] can be obtained in closed form. The expression for [R-2] is a function of several parameters, including the number n of subunits, the fraction f of one kind of subunit, the persistence lengths l(A) and l(B) of the two subunits, and the degree of correlation lambda between successive subunits. The variation of [R-2] with each of these parameters is discussed.