OFFICIAL STATISTICS IN HONG KONG.
4
A
MEMORANDUM ON STUDIES CONDUCTED
WITH A VIEW TO ESTABLISHING A
STATISTICAL DEPARTMENT.
During the course of my leave I devoted some months to the theory of statistics and some months to the practical aspect of governmental statistical work.
As regards the former I was elected a Fellow of the Royal Statistical Society at the end of 1938 and was able to use the Society's Library and to attend their lectures. The principal books on theory which I studied were :—
Raymond Pearl's Medical Biometry and Statistics. Feldman's Biomathematics.
Rugg Statistics as applied to Education.
Bowley Elements of Statistics.
Yule & Kendall Introduction to the Theory of Statistics.
Forsyth Mathematical Analysis of Statistics.
Reitz Handbook of Mathematical Statistics.
Zizek Statistical Averages.
O'Connor Statistics in Theory and Practice. Journal of the Royal Statistical Society.
As well as various mathematical works. As રી certain amount of advanced mathematics is necessary for the theoretical side of statistics it would perhaps be easier if, in future cases, an officer with some mathematical equipment were to be chosen for such a course of study. This purely theoretical side of statistics ought to be of some value in administrative work.
On the practical side I visited various government offices including the statistical departments of the Ministry of Labour, Somerset House, Customs, Board of Trade, Census of Production Office, etc., as well as the London office of the International Labour Office and various firms specializing in statistical equipment and machinery. In Singapore I was given the opportunity of seeing the whole working of their Statistical Department. I was also able to obtain or consult in Canada, America, the Dominions and Colonial Office Library and the Library of the Royal Statistical Society a variety of governmental statistical publications.
It is necessary to draw a sharp distinction between the mathematical theory of statistics and practical statistical work. The mathematical theory, briefly, enables us to produce relatively simple and exact statements in place of complex and com- pendious raw material.
The practical work of statistics consists almost entirely in arranging for the collection of the raw statistics in the most accurate, economical and unambiguous
way.
The two aspects overlap to some extent when it comes to the publication of statistics. The raw material is then summarized and simple statements of its bearing can be made.
The organization of a statistical office is a matter requiring great care because once it has been organized its modus operandi has to be a matter of the most un- varying routine. In fact once the nature of the statistical matter has been settled its form should remain as unalterable as red tape and bureaucracy can make it. If the slightest alteration in this routine is allowed the statistics tend to become-
No comments yet.
Private notes are available after approval.