On inheritance of hair and eye colour / by John Brownlee.

  • Brownlee, John, 1868-1927.
Date:
[1913]
Catalogue details

Licence: In copyright

Credit: On inheritance of hair and eye colour / by John Brownlee. Source: Wellcome Collection.

Provider: This material has been provided by The University of Glasgow Library. The original may be consulted at The University of Glasgow Library.

    19/38 (page 1)
    [Read at the Congress of The Royal Sanitary Institute held at York, July 29th to August 3rd, 1912.] The Theory of Probable Error and its Application to Vital Statistics, by John Brownlee, M.D., D.Sc., Physician Superintendent, City of Glasgow Fever Hospital, Ruchill. WITH the increase of the use of statistics in public health it is becoming increasingly important that an accurate knowledge of the processes by which results are arrived at should be in the hands of all w'orking with figures. The theory of error was originally developed in connection with games of chance, further developed to suit the require- ments of astronomy, and contemporaneously applied from a different point of view to the construction ’ of life tables. In recent years these two applications have converged, till it is now possible to apply many results deduced from the theory of chance to the discussion of problems which could formerly only be attacked by the method of finite differences. 2. Modern mathematical analysis has developed very specially three branches. It has greatly extended the application of the metliod of curve- fitting to smooth observations. It has brought into use a large number of methods for calculating the correlation between different qualities. It has also concerned itself largely with the discussion of probable error. It is this last branch I intend to treat chiefly to-day. 3. This subject falls naturally into four divisions : I. The error due to random selection; II. The assumptions on wdiich the mathematical proofs are based and the modifications required ; III. The influence of experimental error; and IV. The method of testing how far theory and observation agree. I. 4. The subject of probable error due to random sampling is as a rule dismissed in public health text books with a simple statement of Poisson’s Formula, or with a treatment w’hich almost wholly neglects the limitations of its application. The actual mathematics, however, required for its under- standing is not very advanced. The general theorem which is of most importance can be found proved in any elementary text book of Algebi’a, and is as follows. If p be the chance of an event happening and that of it failing to happen so that (/; -f- y) = l, that is, either the event happens or it fails, then in n trials the chance of its happening — )n)