Propagating sharp group homology decompositions. (Q818052)

From ``topological induction theory'', which investigates homology decompositions, one gets, for a fixed collection \(\mathcal C\) of subgroups of \(G\), three different potential induction formulas: by considering the subgroups themselves, their normalizers, or their centralizers. Furthermore, restricting attention to collections of (non-trivial) \(p\)-subgroups of a finite group \(G\), for a fixed prime \(p\), there are at least nine different collections commonly studied, which thus give some 27 potential induction formulas. In this paper the authors give a systematic study of the interrelationship among these 27 potential formulas and they are able to settle, either in the positive or the negative, which of these 27 formulas give rise to ``sharp homology decompositions'' [\textit{W. G. Dwyer}, Proc. Symp. Pure Math. 63, 197-220 (1998; Zbl 0891.55021)] -- recovering many previous results, as well as closing the gaps in the existing literature.