Ackerman performs a cross-cultural analysis on the structural determinants of divorce rate as originally hypothesized by Max Gluckman and elaborated on by other researchers. Ackerman's results suggest that when spouses share a network of affiliation, divorce rates are low; when spouses maintain separate affiliations, divorce rates are high. Ackerman's statistical analysis and discussion provide an explanatory framework for further research.

Sampling 62 societies from the HRAF database (32 bilateral, 21 patrilineal, 6 matrilineal, and 3 double-unilineal societies)the author asks whether divorce rates can be predicted by the descent systems or the network of marriage affiliations. Findings do not support the theory that divorce rates will be predicted by descent (patrilineal vs bilateral) - but they are more consistent with the idea that conjunctive affiliations are predictive of low divorce rates in bilateral societies. In lineal societies the levirate is used as an indicator.