Model Tables: the Coale-Demeny set

This system of model life tables is presented as four families:

They are based on 326 pairs of life tables, mostly from developed countries – overwhelmingly Europe – and some of these dating from the 19th century. None are from Africa. The compass indicators therefore reflect, surprisingly, not four quarters of the globe but largely the mortality characteristics of different parts of Europe! This raises an interesting demographic conundrum – that the name of the pattern is largely irrelevant to its geographic application!

This is an anomaly that confuses non-demographers but is important to understand. The use of model life tables is entirely pragmatic. The patterns are not required to be applied exclusively to the areas from which they were derived – the Coale Demeny patterns are used worldwide despite the fact that they are largely based on European sources. The thinking behind this is that it is more important to have reliable sources giving patterns of mortality that are known to exist, rather than have geographically specific patterns based on less reliable data.

The naming of mortality patterns have caused much confusion but they should be applied pragmatically – use the pattern that best fits your data, not the name that seems most appropriate!

The North life table was fitted to (mostly) Scandinavian mortality data, while the South family of life tables were derived (mostly) from Mediterranean European populations and the East tables mostly from Eastern European sources.

The West family is not an “all-in” pattern but a pattern that embraces mortality data largely from life tables in Western Europe but also a mixed bag of other sources outside of Europe. This pattern brings together some quite disparate tables and is therefore regarded as a generic pattern which can be used as a default if the three other patterns are not suitable.

The four families represent age patterns of mortality that differ primarily in the relationship of child to adult mortality.