In this paper two approaches were presented for dealing with multivariate analysis of life tables. The need for this type of analysis became obvious after individual level data became available by the recent WFS surveys and because of the growing concern for regarding fertility as a sequential process rather than looking at its final outcome. The analysis has become possible by the availability of several computer packages analysis. The first approach uses a hazard model in which the risk of transition allowed to vary among women of different higher parity is characteristics at different durations. The second treats every duration segment separately and uses a log linear model to describe the relationship between the proportion of women having a birth within the segment and the women's characteristics. Several references are given to the statistical background material on the methods as well as to some substantive demographic research in which these methods were applied. The choice of a particular method depends on the type of the available data, and - to a large extent -on the availability of suitable computer software.