Life-tables for Grenada and
Trinidad and Tobago, 1891-1921
The derivation of the net reproduction rate for both territories meant that life‐ tables for the female population were necessary. In the case of Grenada, none existed. In the case of Trinidad and Tobago life-tables existed only from 1901 onwards. The non-existence of tables for years prior to this is due to the fact that there are no data on deaths by age for the population of Tobago prior to this date. Furthermore, the vital statistics of the territories were not treated as one until after 1891, the census year. Nevertheless, a life-table based on deaths in Trinidad's population alone was constructed for 1891. The other life-tables for the census years 1901, 1911, and 1921 were taken from G. W. Roberts and J. Harewood's "Lifetables for West Indian Populations 1945-47 and 1950‐ 52", Census Research Programme, no. 14.
No life-tables previously existed for Grenada. Life-tables were constructed for the census years 1891,1901,1911 and 1921. In the case of these life-tables and that for Trinidad of 1891, they were based on deaths for three year periods around the time of the census; the empirical population at risk of dying being derived from the censuses of the respective years.
The infant years were divided as follows: 0, 1-4, 5-10. The following formula was utilized in the conversion of the Mx, the rate of dying in the empirical population to qx, the probability of dying in the life-table population: nqx = n · Mx / 1 + n (1 - nax) · nMx where ax varied in value according to age group. For age 0, it assumed a value of 0.3. For the age category 1-4, it assumed a value of 0.4. From age 5 onwards it assumed a value of 0.5. The other columns were constructed using standard life-table construction procedure. The life‐ tables are reproduced below.