The problem of calculating statistical tensors p;(Z) for the nuclear-ordered state in enhanced compounds such as HoVO,, PrNi, etc, is examined within the framework of both a simple effective-field model and more sophisticated many-body theories, such as the RPA. It is shown that for simple ferromagnets and the anti-ferromagnets, a result, originally due to Callen and Shtrikman, can be used to show that there is a one-to-one relationship between the p!$Z) calculated using many-body Green function theories and the simple effective-field model. In addition a new recursion relationship is derived, linking statistical tensors of differing rank. This relationship holds for a wide class of renormalised collective excitation theories and is far more convenient to use in practice than either the momentgenerating function of Callen or the methods of Tahir-Kheli and ter Haar.Using first-order perturbation theory it is also shown that the Callen and Shtrikman result breaks down in the presence of a small axially symmetric quadrupole interaction. Thus it is unlikely that a simple check on the accuracy of 'simple effective-field statistical tensors' can be devised in the spirit of the Callen and Shtrikman result. Some comments are also made concerning the nuclear statistical tensors used to describe the y-ray emission patterns from the enhanced nuclear magnet HoVO, (Allsop er al).