View report page

All the correlations are significant. The indices of density in 1950-52 were not so
highly correlated with the birth rate as in former years, and the correlation between
birth rate and the proportion in the lowest social classes of the population has also
decreased since 1920-22, though the difference between 0.811 and 0.670 is not significant.
Infant mortality
Hersch (1943) chose infant mortality as the best single numerical index of the
degree of civilization of a population. He considered that a decrease in this rate reflected
the advances made in hygiene, medicine, chemistry, and public instruction, and that it
was the best sign of improved living conditions. When Hersch was writing, infant
mortality in Europe ranged from 37 in the Netherlands to 180 in Rumania, and even
higher rates occurred in Eastern countries.
In England and Wales infant mortality has been used in the past as an index of the
socio-economic level of the community, since the highest rates were found to occur
where living conditions were worst, and infant mortality varied directly with social
class. McKinlay (1928) attempted to standardize for varying social conditions in the
London boroughs, but found that such standardization for economic status did not
reduce the variability between the boroughs by as much as seemed a priori probable.
Stocks (1928), using the occupations of the males from the census returns, attempted
to find a correction factor which would reflect the social make-up of the London
boroughs. This correction, however, only reduced the variability between the infant
mortality rates of the London boroughs from 17 to 14 per cent. in 1911-13, and from
15 to 13 per cent. in 1921-25. The findings of McKinlay and Stocks are in agreement,
and clearly the economic factor, as measured by their indices, was not very important
in determining the level of infant mortality in the various boroughs. During the past
25 years the effect of many variables has been studied : e.g. indices of density, proportion
of mothers employed, fertility rate, and proportion of males in the professional classes
and in the lowest type of labour, etc. These investigations have shown that infant
mortality was affected by socio-economic conditions, but since the variables used were
themselves inter-related it has not been possible to assess the relative importance of
their contributions. In recent years infant mortality has fallen rapidly. In England and
Wales the rate was 154 deaths per 1,000 live births in 1900, by 1922 this had been halved,
and in 1930 the rate was down to 60. Between 1930 and 1941 the rate fluctuated between
51 and 66, but since 1941 when it was 60 it has fallen steadily to 25.5 in 1954.
In 1950-52 the infant mortality rate for London (24) was below the rate for the whole
country (29), and much below the rates for the county boroughs of Durham (40),
Lancashire (36), Staffordshire (34), and Yorkshire West Riding (31). Not only was the
infant mortality lower in London than in the large industrial towns of the north but
172