Memoranda presented to the Royal Commission.

  • Great Britain. Royal Commission on Population
Date:
1950
Catalogue details

Licence: Public Domain Mark

Credit: Memoranda presented to the Royal Commission. Source: Wellcome Collection.

    48/128 page 42
    regression of Otis I.B. on Binet I.Q. is used, b = 2.026859 (p. 192), the value is 2.0418 I.Q. in each generation. If, as appears to be generally agreed, the Binet I.Q. test is more accurate than the Otis I.B., the latter value will be more proper. If Fraser Roberts gives the figure 1.5 ascribed to him by Thomson (which I have not been able to verify) it is somewhat an understatement of what his data indicate. There is in any case no reason a priori for supposing that the rate of decline of intelligence is neces- sarily the same in Leicester and in Bath. For reasons which both authors give, the value of about 2 points I.Q. per generation obtained at Bath and of 3 points per generation at Leicester are doubtless both appreciably under- estimates. A repetition of the same tests with approximately the same numbers at Bath, after a generation, e.g., in 1965, would doubtless show a significant change, in the average intelligence though it would not estimate it accur- ately, since the standard error of the comparison would be about 0.8 units I.Q. per generation. Such a direct test would include the influence of those leaving no children. In the meanwhile tests closely conforming with that of Fraser Roberts at Bath should be carried out in a number of larger towns. August, 1944. 3. (@) COMMENT BY PROFESSOR J. B. S. HALDANE [ have read Professor Thomson’s memorandum with great interest. His main thesis seems to be fairly well established. Nevertheless there are certain facts which suggest that some of his conclusions might be modified. Where an attempt is made, within a social class, to correlate fertility with parental intelligence, positive correlations are often found. For example, Phillips(*) graded 1899 Harvard graduates of 1899-1901 according to their success and fertility, Class I being eminent, and Class V social failures. The results were as follows :— Class 1 Z 3 4 5 Mean number of children Ae 2-19 1-93 1-67 1A 0:80 The gradient was partly, but not wholly, due to differences in marriage rate. Huntington and Whitney(’) found similar results for Yale. This sug- gests that if the correlations noted by Thomson imply a negative correla- tion between parental intelligence and fertility, this may be wholly or mainly due to differences of fertility between different social classes, rather than to differences within a class. The conclusion that the average innate intelligence of the British popu- lation is falling (in so far as a meaning can be attached to such a phrase) is also very possibly true, but does not follow from the data presented. To take an extreme example, “eversporting” stocks (Matthiola incana) have single flowers, but over 50 per cent. of their progeny have double flowers, and are therefore completely sterile. Nevertheless the percentage of sterile progeny does not diminish from one generation to another. This is due to a selective process among pollen grains acting in the contrary direction. Even in existing society there is considerable selection against mental defect, defectives having a higher death rate and less fertility among those (‘) Phillips, J.C. (1927). Success and the Birthrate. Harvard Graduates’ Magazine xxxv, 565. (7) Huntington, E., and Whitney, L. F. (1927). The Builders of America. (New York).
    No text description is available for this image
    No text description is available for this image
    No text description is available for this image