the natural history of population.by raymond pearl

4
The Natural History of Population. by Raymond Pearl Review by: E. C. R. Journal of the Royal Statistical Society, Vol. 102, No. 3 (1939), pp. 452-454 Published by: Wiley for the Royal Statistical Society Stable URL: http://www.jstor.org/stable/2980071 . Accessed: 25/06/2014 06:17 Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp . JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected]. . Wiley and Royal Statistical Society are collaborating with JSTOR to digitize, preserve and extend access to Journal of the Royal Statistical Society. http://www.jstor.org This content downloaded from 195.78.108.81 on Wed, 25 Jun 2014 06:17:41 AM All use subject to JSTOR Terms and Conditions

Upload: review-by-e-c-r

Post on 30-Jan-2017

212 views

Category:

Documents


0 download

TRANSCRIPT

The Natural History of Population. by Raymond PearlReview by: E. C. R.Journal of the Royal Statistical Society, Vol. 102, No. 3 (1939), pp. 452-454Published by: Wiley for the Royal Statistical SocietyStable URL: http://www.jstor.org/stable/2980071 .

Accessed: 25/06/2014 06:17

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Wiley and Royal Statistical Society are collaborating with JSTOR to digitize, preserve and extend access toJournal of the Royal Statistical Society.

http://www.jstor.org

This content downloaded from 195.78.108.81 on Wed, 25 Jun 2014 06:17:41 AMAll use subject to JSTOR Terms and Conditions

452 Reviews of Statistical and Economic Books [Part III,

statistics is of the cum hoc ergo propter hoc type, for the analysis and exposal of which partial correlations, or the ideas on which they are based, are essential. A further chapter on the logic of partial correlation analysis would add greatly to the value of the book.

The translation has been faithfully done; in fact, in places a little too faithfully for comfortable reading. Thirty or forty definite articles sprinkled through the book would improve such sentences as " In older literature we occasionally find the value of standard error of empirical correlation coefficient denoted by . . . . etc." There are some unaccountable omissions of references appearing in the German edition, particularly on page 192 in regard to some of the fundamental papers on the distribution of the correlation co- efficient in samples from an infinite normal universe. But these are venial faults compared with the omission of the subject-index- a form of crime in translation which appears to be on the increase and deserves the sternest penalties. M. G. K.

3.-A n Introduction to Modern Statistical Methods. By Paul R. Rider. Chapman and Hall. 8-" x 5k". ix + 220 pp. I3S. 6d.

The object of this book is to give an account of the most widely used statistical methods which have been invented in the last twenty or thirty years. The author also claims that since the earlier chapters develop the fundamental concepts of statistics the book is suitable as a text-book for a first course in the subject.

Excluding the exercises at the end of each chapter, 7 pages are devoted to frequency distributions, 13 to averages and moments, i6 to regression, including multiple and curvilinear regression, and 15 to correlation. These four are presumably the ones which develop the fundamental concepts. Then follow chapters on the binomial and normal distributions, the t-distribution, x2, the analysis of variance, and experimental design.

There is a tendency for books of this type, which frankly abandon any attempt to discuss the theoretical bases of the methods put forward, to degenerate into a series of recipes. Mr. Rider's book seems open to criticism on these lines. Nearly all the tests with which it deals are based on normality in the parent universe, but nowhere does the author give any indication how to decide whether an assump- tion of normality is legitimate in any given case. Nor does he discuss sampling, randomness, or probability. Even the term "standard error " is not defined or used. It wonild seem that Mr. Rider has over-compressed and over-simplified the book. M. G. K.

4.-The Natural History of Population. By Raymond Pearl. Johns Hopkins University; Oxford University Press, 19&9, xii + 416 pp. Ios. 6d.

The Heath Clark Lectures, University of London, were delivered in 1937 by Professor Raymond Pearl, one of the Royal Statistical Society's distinguished Honorary Fellows, and the present book gives the material which the lectures summarized. Professor Pearl's position as one of the world's leading authorities on all sides

This content downloaded from 195.78.108.81 on Wed, 25 Jun 2014 06:17:41 AMAll use subject to JSTOR Terms and Conditions

1939] Reviews of Statistical and Economic Books 453

of the so-called " Population Problem " needs no stressing. He has personally been responsible for many and varied publications in this field, and the total volume of work which has emanated from people in his department at the Johns Hopkins Universityis probably only known to that department. In all his work. there has been obviously a desire to get to the bottom of the particular subject. This motif runs all through the present book. He analyses the problem, showing what information is necessary before any judgment may be given, and then he presents the results of his researches and of his investigation into the work of others. An illustration of this will suffice. At the beginning of the book he is desirous of giving information relating to the age at menarche. He examines masses of literature on the subject and concludes " as a statistical generaliza- tion, the average age of menarche as judged by a world-wide sample of considerable size, is not far from 15 years." This average is well known to many of us, not from statistical investigation, but from vague statements culled from scientific or semi-scientific literature. But Pearl gives us the other elements, besides the average. We now know something of the variability of this statistic, what its lowest limit is and how it varies geographically.

Similarly, when he tackles the problem of measuring fertility, he takes into consideration all the various factors which are likely to be involved. He presents a chart (p. 96), showing the interplay of the various factors, economic circumstances, education, age, specific innate reproductive capacity, density of population, physical health, reproductive wastage rate, and so on. All this is necessary if we are to compare statistical results of one group with those of another.

He concludes (p. 164), after an exhaustive analysis which takes account of the various factors, that the proportion of women physio- logically capable of reproduction who actually reproduced in 1930 was smaller for all the age groups of mothers, except the youngest, than it was in 1920. He asserts " that there was some reason to believe that this change probably could not be wholly attributed to an increase of contraceptive efforts in the population in the ten-year period."

Having argued the need for records of all the facts relative to human fertility, he gives an account of a mammoth investigation planned by him with the aid of funds from the Milbank Memorial Fund, where information relating to marital history was sought from patients in hospitals in a number of American cities. In the book he gives us considerable information about this important investiga- tion, and makes us realize its magnitude and its scope. Ultimately, what was sought was the difference between the reproductive patterns of females who assert that they never practised contraceptioif and those who assert that sometime during their reproductive period they have been contraceptors. He realizes fully the difficulties in- volved in grouping together all those persons who have practised contraception of various kinds in varying degrees, and he knows the inherent weaknesses of his facts. But the main conclusion showing the differences that exist between the various groups in his analysis

This content downloaded from 195.78.108.81 on Wed, 25 Jun 2014 06:17:41 AMAll use subject to JSTOR Terms and Conditions

454 Reviews of Statistical and Economic Books [Part III,

is enough to give some indication, at any rate in default of better information, of the effect of the recorded practice of contraception.

He concludes " that contraception, as actually practised in the population, is not having nearly so great an effect in lowering the reproductivity . . . as many would have us believe " (p. 160). He also concludes, from an analysis involving the data grouped in economic, educational, and religious classes, that " if it were not for the effect of contraceptive efforts and the practice of criminal abortion, together with correlated habits as to postponement of marriage, there would apparently be little or no significant differential fertility as between economic, educational and religious classes of urban American married couples." In other words, the differences are due to differences in contraceptive efforts and not to biological differences.

The first part of the book is concerned with many biological facts relating to populations, especially to fertility, the middle of the book is devoted to a discussion of the large investigation on contraceptive habits already referred to, and the last chapter is on the subject of world population. Here, in comparatively few pages, he presents a picture of the growth of population through the ages and shows its distribution by area and by "empires". Some of his opinions in this chapter are necessarily person. He refers to the growth and decay of former empires. He seems to think that man accepts calmly the prospect of the inevitable decline of the present civilization.

There is obviously some difficulty involved in the presentation of the results of scientific work of this sort, especially when it brings in.statistics. Prof. Pearl has a simplicity of expression which rides through the use of long words and scientific expression. At times, also, the man who obviously delights in paradox and has a strong sense of humour breaks through, and gives us, on p. 185, in reference to a statistical table, " A reasonable way of putting the matter is that while more highly educated girls may or may not be more moral than their educationally less fortunate sisters, at least they know better, on the average, how to forefend the dire consequences of such excursions as they may choose to make into the realm of amatory dalliance."

The book will occupy a prominent position in the literature de- voted to the scientific study of population problems. E. C. R.

5.-Aperqu de la De'mographie des Divers Pays du Monde 1929- 1936. The Hague: Office Permanent de l'Institut International de Statistique, 1939. 102" x 74". 433 pp. Fl. 8.

All students who have occasion to use vital and census statistics will be glad that a new volume of the Aperqu De6mographiq4e has at long last appeared. The new volume is, moreover, a c nsi1erable improvement on the 1929 issue. It contains detailed bibliographic references and a much wider range of tables than earlier volumes. Thus, for example, there are not only expectations of life, but also life-table mortality coefficients and survivor tables. Life-table

This content downloaded from 195.78.108.81 on Wed, 25 Jun 2014 06:17:41 AMAll use subject to JSTOR Terms and Conditions