Reading the tribute published by colleagues at Berkeley for the statistician William Leonard Crum, one would come away with the impression that he was not just a nice man and good colleague, but that he was even a pretty, pretty good statistician. In fact Crum lagged the frontier of mathematical statistics by at least a full generation, cf. the 1945 report written by Crum and his younger colleague Edward Frickey on Harvard’s statistics and national income courses.
I stumbled upon a reprint of the short 1926 article by Crum, transcribed below, at the Library of Congress, Manuscripts Division, Francis Willcox papers, 1851-1961 (Box 40, Folder “Outlines of Economics Statistics”). What the article lacks in profundity, the reprint provides in a somewhat better image of Crum in his early ’30s that I use here rather than a screen shot of the hathitrust.org image that provides an internet source for the text.
____________________________
In Memorium
University of California, 1968
William Leonard Crum,
Economics; Business Administration: Berkeley
1894-1967
Professor of Economics, EmeritusOn June 11, 1959, the University of California on the Berkeley campus conferred the honorary degree Doctor of Laws upon Professor W. L. Crum. The citation read as follows:
“Your academic career has involved membership in faculties of four great universities: Yale, Harvard, Stanford, and California. A warmly respected colleague, valued teacher, and careful administrator, an economist who brings to the ‘dismal science’ a wealth of human understanding, judgment, and objectivity…”
Professor Crum was born on April 8, 1894, in Hoosick Falls, New York. He was graduated from Williams College in 1914 in Physics and received the M.A. and Ph.D. degrees from Yale in 1916 and 1917 in Mathematics. Following his service in World War I, which included 13 months’ duty in the artillery in France, he returned to Yale’s Mathematics faculty. Beginning in 1923, he served on the faculty of Economics at Harvard for 25 years except for two years at Stanford from 1927-29 as Professor of Statistics. He came to
Berkeley in 1948 as Professor of Economics in the Departments of Business Administration and Economics. Honorary degrees were conferred upon him by Williams College (1940), Harvard (1945), and California (1959).
Leonard Crum had one of the finest minds and best tool kits (science, mathematics, statistics, economic theory) in the economics profession. Beginning in 1918, there was a steady flow of books, monographs, and scholarly papers both in the refinement and application of mathematical, statistical methods and in substantive contributions. Early in his career he became a recognized pioneering authority in business cycle analysis. From 1923 to 1935 he was associated with the new Harvard Economic Society in various capacities, including the posts of Vice President and President. From 1936 to 1949 he was a Director of the National Bureau of Economic Research. Prior to World War II he was consultant, Division of Research and Statistics and Division of Tax Research, United States Treasury. He has served many journals in an editorial capacity, including Harvard’s Review of Economic Statistics and Quarterly Journal of Economics. A number of his early publications were classics in their fields—among them, Advertising Fluctuations, Seasonal and Cyclical (1927), Corporate Earning Power (1929), and Corporate Size and Earning Power (1939). These early studies provided the background for his important The Age Structure of the Corporate System (1953), published by the University of California Press. In Berkeley his primary teaching field was private finance and investments, in which he was a leading authority.Leonard Crum was an exceedingly active and influential member of the Berkeley faculties in Business Administration and in Economics. His contributions derived not only from his continuous leadership and productivity as a scholar and a teacher but also from conscientious, effective participation and leadership in all aspects of departmental, school, campus, and University-wide responsibilities. His budgetary requests and reports as Vice-Chairman of the Department of Business Administration became the models for all other administrative units. He always had time, or made time, to serve on faculty and administrative committees, to appraise the research papers of colleagues and students, to talk over personal investment problems, to serve as member or chairman of campus or University-wide ad hoc committees.
At the memorial service on June 1, 1967, a colleague in another department remarked that Leonard Crum, time and again, was made chairman of committees when there were deep divisions of opinion since “he alone could find the consensus, and compel its acceptance.” He also noted, “Few combine Leonard’s proud independence of judgement with his ability to find a basis for common action in the service of the University.” Yet it was also noted by a colleague in the School of Business Administration that Dr. Crum “usually offered advice with such warmth and diffidence that most of us believed that we had thought up the ideas ourselves…. He was an individualist, who valued and heeded the views of others, but he did not shape his own ideas by waiting for the results of public opinion polls.” Another colleague from his own faculty remarked upon his fairness and forthrightness and his sense of humor and dry wit.
His basic kindliness, warmth of personality, and warm camaraderie were also stressed, together with an inherent New England formality, dignity, and reticence in manner. Most important of all, however, was his basic integrity, which was reflected in all aspects of his own work and in his administrative and professional relations.
All will agree with the observation also made at the Memorial Service that “he was a very uncommon man, indeed,” a man of extraordinary talents and highest standards of performance and almost complete disinterest in recognition for himself alone. Consequently his personal leadership and influence were unquestioned.
Leonard Crum was married on June 29, 1938, to Eleanor Marshall Evans who survives him.
E. T. Grether D. Votaw P. F. Wendt
Source: University of California, In Memorium, 1968, pp. 32-34.
____________________________
THE TEACHING OF ECONOMIC AND BUSINESS STATISTICS
By WILLIAM L. CRUM *
[*Co-author, with A. C. Patton, of An Introduction to the Methods of Economic Statistics, published by A. W. Shaw Company, 1925]
In recent years and especially following the war, there has been an enormous increase in the attention devoted to the elementary methods of statistical analysis. Not only in schools of business administration but also in the economics departments of liberal colleges there has been a vigorous development in the teaching of statistics, particularly as applied to economic and business problems. This tendency has resulted, in part, from the realization during the war of the great utility of statistical records and their interpretation, in the approach to and solution of the business problems of private enterprise, and the economic problems of public affairs. It has resulted also in considerable measure from the great interest in forecasting the cycles of business prosperity and depression. Such forecasting has been advanced recently by the application of statistical methods to the analysis of the already available and rapidly accumulating records of economic and business conditions. No doubt also the increasing attention to economies in production and distribution, which became necessary during the readjustment following war-time inflation, has played a large part in expanding the emphasis upon the analysis of numerical records of performance.
The teaching problems involved in presenting a course in the methods of business statistics are somewhat more specialized than those incident to a course in economic statistics. In teaching economic statistics it is now rather generally understood that certain well-defined fundamental concepts must be introduced at an early stage. These relate to the nature of statistical facts, the standard methods for presenting statistical data, and the elementary arithmetical devices for summarizing such data. For the most part, these introductory matters comprise a discussion of the collection and assembling of statistical material, the organization of such material into tables or charts, and the derivation from such material of those particular statistical numbers which serve to give a brief but adequate picture of the main properties of individual groups of data. Indeed, the chief teaching problem here consists in the development of methods of deriving summary statistical numbers — numbers concerned primarily with the averages and with measures of dispersion or variability — of the particular series of economic data which may be under examination.
It is here first that the teacher is confronted with the question of the mathematical knowledge to be assumed. Although it is undoubtedly much simpler to present the essential principles of statistical analysis to students or readers who have a working understanding of college mathematics, it is an important fact that at present the majority of those who desire to secure an adequate knowledge of economic and business statistics do not have this mathematical equipment. It is in practice necessary, therefore, to outline the course with a view to making it understandable to this majority of students. An examination of the text-books which have been written on economic and business statistics in recent years clearly reveals the tendency, at least in the majority of instances, to give a treatment which can be comprehended by those individuals who have only a minimum of mathematical knowledge.
When once the basic concepts essential to an understanding of statistical analysis have been grasped, the student is ready for the study of numerous specific economic applications of statistics. He can be introduced, for example, to the general problems of price index numbers, of individual incomes, of foreign trade movements, of manufacturing and basic production, of the business cycle and its forecasting, and to other economic problems which are currently approached from the statistical side. Presumably, the bulk of effort in a course in economic statistics is given to these applications, to the adaptation of the elementary concepts to the particular needs incident to these applications, and to the development of certain specialized methods which are required in many particular applications. A satisfactory course of this sort not only teaches the student how to analyse the statistics involved in the economic applications, but it gives him also some schooling in the interpretation of his results, and it cautions him emphatically against the too confident inferences which he is likely to draw if he disregards the necessary limitations on the statistical processes.
The teaching of business statistics is necessarily somewhat more specialized than the teaching of economic statistics. It may fairly be said that the student of business statistics must first have a thorough course in the methods of economic statistics. Although it is possible for him to cut short some of the extensive study usually given in economic statistics to the applications of statistical method to specific economic problems, it is essential that he have a working knowledge of the basic concepts before attempting to make any considerable study of special business statistics. To be sure, it is quite feasible for the course in business statistics to confine its illustrations of method, even of elementary method, to business data rather than economic data, and in this way to introduce the student early to those special and frequently limited problems which confront the individual enterprise rather than industry or the economic structure as a whole.
When once the student of business statistics has gone over the preliminary ground which is also covered in economic statistics, and has become acquainted with some of those economic applications which are of importance also in business statistics — because a considerable part of business statistics, sometimes called external statistics, includes a large portion of so-called economic statistics — he is ready to go forward with the peculiar problems of business statistics. These problems are frequently so highly specialized that it is very difficult to outline a complete course based upon a few general principles of analysis, as is the case in economic statistics. Here, rather, one is impelled to a treatment somewhat resembling the so-called case system of instruction, in which the student takes up one particular problem after another and learns by a study of individual and frequently very different “situations” those principles and practices which fit him to cope with the ordinary problems of business statistics which he may encounter in the administrative work of a particular organization.
In conclusion, then, it may be said that a course in economic or business statistics commences with a study of elementary principles and methods common to both subjects and that in either case considerable attention must be given to the application of statistical methods to general economic problems. At some stage, however, the course in business statistics must break off from the general outline of the course in economic statistics, and from there on must be concerned largely with the treatment of numerous and diverse special problems each presenting its own difficulties and requiring its own peculiar method of treatment. In both courses a large amount of actual routine technical work on the part of the student is essential if he is to learn the subject thoroughly, and in both courses the teacher must lay a great deal of emphasis upon the interpretation of the results found by analysis.
Source: Ex Libris. Vol. I, (May 1926), No. 3, pp. 14-16. “A journal devoted to the professional interests of business, agriculture, engineering, and the social sciences.”
Image Source: Portrait of William L. Crum, ibid., p. 15.