摘要:

SummaryPower and limitations of statisticsIt is certainly worth‐while to devote some attention to the power of statistics, which from the point of view of both science and technique is showing a steady development. This is illustrated, for example, by the considerable interest displayed in the work of the Association for Statistics, as is evident from the growing number of members, the number of candidates for the examinations for statistical analyst and the number of people attending the Statistical Days. Furthermore, there is the appreciation shown by industry of the productivity investigations carried out by the Central Bureau of Statistics and of the establishment of the Quality Control Service for Industry.Naturally, it is in itself commendable that efforts are being made to analyse and gauge social life. One cannot escape the impression, however, that in the case of the Netherlands—taking the authorities, industry and social research as a whole—we are aiming rather higher than can be achieved with the intellectual man power available. In this connection the speaker pointed out that the people of the Netherlands will have to give serious attention both to the selection and training of the talented and to the manner in which the most efficient use can be made of the talent present among the population. As matters stand now we are incapable of realizing our power.The power of statistics is, moreover, evident from the part it is beginning to play in government policy. We have nowadays a whole series of index figures of the Central Bureau of Statistics, which fulfil a basic function in determining the economic and social policies to be followed. At the present time, however, the influence of statistics often goes much further. Its function can be compared to that of the “automatic pilot”, i.e. any change in a given statistical magnitude involves certain consequences in virtue of some law, agreement or contract. The speaker illustrated this point with a number of examples. He expressed his appreciation of the fact that there is a general tendency to take statistics seriously. There is no one more aware of his limitations than the statistician who is promoted to the position of oracle. His statistical series were originally compiled for purposes other than those for which they are now used. According to the speaker it was doubtful whether the statistician would ever be able to come up fully to the expectations which people nou ‘ jve of him. In all statistics there is invariably an element of abstraction and generalization. The reality is always infinitely more differentiated than can he expressed by means of statistics. Anyone handling the index figure for the cost of domestic consumption, for example, should know that it is based on an assumption in respect of reality, viz. that this consumption is always the same, and not on reality itself. This does not detract from the value of the index figures as a theoretical aid, but as a basis for practical work they should be surrounded with a multitude of warning signs.In conclusion the speaker pointed out a more popular example of the power of statistics, viz. the poll whereby the trend of opinion is established. In this sphere, too, the limits of statistics are clearly defined. In how far is it possible to determine statistically certain views, ideas, attitudes, etc. There are innumerable magnitudes of great social importance in the life of a community, which cannot be classified and thus cannot be measured st

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1953.tb00999.x

出版商:Blackwell Publishing Ltd    

年代:1953

数据来源: WILEY

摘要:

SummaryObservation and Rational Behaviour.Wald's theory of Decision Functions is illustrated by means of an example, using D. J. Bross' terminology. This paper is meant to supplement prof. D. van Dantzig's 1952 paper, entitled “Playing against Nature”, also read before the Vereniging voor Statist

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1953.tb01000.x

出版商:Blackwell Publishing Ltd    

年代:1953

数据来源: WILEY

摘要:

SummaryIn this lecture the responsibilities of the statistician are discussed. He always has to find a compromise between his mathematical and his social conscience, because he hardly ever would be allowed to apply the mathematical theory to practical problems if he would consult his mathematical conscience only. This is explained by an example in which his research serves for preparing the important government decisions to be taken in order to prevent future flood calamities like the one that happened on February 1st,1953.

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1953.tb01001.x

出版商:Blackwell Publishing Ltd    

年代:1953

数据来源: WILEY

摘要:

SummaryQuantification in the social sciencesAfter some remarks about the psychological background of quantification, four kinds of entities are distinguished: measurable, scalable, modal or classifiable variables, and finally variables that cannot even be classifiedDifficulties of characterizing entities with respect quantifiability are exemplified by the notion of utility in economics, and to toleration in sociology.Practical problems of measurement in the social sciences are traced back to man's ability to lie and his dis inclination to speak, when questioned about matters of importance to him.The limitations of summarizing measures, such as totals and averages, are brought to the fore, while the neglect of variances in social sciences is criticized.Because the mathemical form of economic and social relationships is rarely specified by theory, measures such as coefficients of elasticity cannot be determined with precision. Moreover, the latter are not really constants, since time cannot be explained away altogether.

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1953.tb01002.x

出版商:Blackwell Publishing Ltd    

年代:1953

数据来源: WILEY

摘要:

SummaryQuantification in Medical ResearchIn medecine statistical methods can be used in the comparison of efficiency of several treatments. In the past in most cases intuition was used to come to a judgment on the results of the comparison. Quantification of the results of treatments is made possible by the application of statistical methods, and so the exactness of judgments can be enlarged. For this reason we may speak of the power of statistics in this field. Some examples of this power are given.The application of statistical methods in comparing results of treatments bring along some ethical problems of which both physician and statistician ought to be aware. It is necessary that observations in patients are used as efficiently as possible!

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1953.tb01003.x

出版商:Blackwell Publishing Ltd    

年代:1953

数据来源: WILEY

摘要:

SummaryQuantification in language and Crypto‐analysisIn this paper the statistical properties of different languages are considered and it is shown howevery language may be charactarised by a parameter, the so called “invariant paramete”. This parameter, and the character of a language in general, can be changed by crypto‐transformations. Statistical methods for finding the key of crypto‐transformations are discussed and illustrated by examples. It is shown how these methods can be made ineffectual by using special crypto‐transformations, eradicating those statistical properties of the language, which constitute systematic deviations from randomness in the numbers and combinations of l

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1953.tb01004.x

出版商:Blackwell Publishing Ltd    

年代:1953

数据来源: WILEY

ISSN:0039-0402    

DOI:10.1111/j.1467-9574.1953.tb01005.x

出版商:Blackwell Publishing Ltd    

年代:1953

数据来源: WILEY