In this paper we will deal with the problem of the verisimilitude of qualitative
theories stated in first order languages with two or more families of predicates (Qtheories).
In particular, we will explicate the notion of Q-theory and define suitable
measures of the qualitative verisimilitude of Q-theories. Afterwards, we will introduce
appropriate measures for the statistical verisimilitude of Q-theories. This will allow us to
show that Q-theories can be used to describe the statistical structure of cross classified
populations, and that their adequacy in this task can be evaluated by measuring their
statistical verisimilitude.