This paper is a consideration of the possibility of combining sampling procedures and role theory as methods of social research. It is one contention of this study that these are currently the chief centres of interest in sociological theorizing. The first, shared by several other sciences, is part of a broader interest in statistical procedures. The other is an integral part of the renewed interest in the study of social organization as such.

On the surface it might appear that these two interests are complementary in character, and that both would share in the attention of persons inclined to considerations of theory. Actually the situation is quite otherwise. Students in the two fields have remained isolated from each other to an eminent degree. The intent of this paper is to attempt a bridge between these two interests, by exploring the possibility of extending the techniques of sampling to the analysis of social organization.

The paper is divided into three parts. The first section is a very abbreviated introduction to some general features of sampling procedures and of the nature of social organization. The second part is a consideration of a specific study in each of the two fields. The final section deals with the application of both types of approach to a research situation.

A chemist who either discussed his subject, or worked in his laboratory, on the assumption that all elements were equal, or that matter was composed of single homogeneous units called “molecules,” would not remain long in his profession. Force of ridicule, or of some even more drastic happening, would remove him. But the economics of many practical and pressing problems is continually bedevilled by discussions, and even on occasion policies, which involve a sort of stubborn simplification quite unworried by the facts. It is certainly understandable that, in the field of housing, this complaint has arisen frequently from those who have tried to deal with the subject from a social rather than a particularist viewpoint, and in its full dimensions rather than within the limits of some familiar phase. In the effort to get low-rent housing into its proper perspective, there are two obstacles which are particularly troublesome. One is the assumption that all housing “units” are the same, the kind of view which leads to the belief that “one thousand houses built” is the same contribution to national need, no matter what kind of houses they may be. The other is the argument or belief which really involves the assumption that peoples' incomes are broadly equal, and all comfortably above the average. The most frequent version is that most people, or at least a majority, can afford to own their own home, or should be encouraged to do so. A more valid but troublesome one in its “all-or-nothing” form is that cheapening the cost of construction is the “real” problem; and there are other variants. Another version, less frequently heard in these days of pressure, is that people live in slums because they like it.

World events during the past twenty years, the great depression and the war, have challenged economists to improve and enlarge their theoretical framework. The stresses and strains of this period have brought forward vitally important problems which demanded solution. At the same time, these events helped to produce an environment which responded more generously to the social scientist's quest for new data. There can be little doubt that there was an intimate interdependence between developments in economic theory and analysis, and the growth and systematization of quantitative data.

Perhaps the most significant of these developments has been in the field of national accounts. The purpose of this paper is briefly to describe and illustrate this development in Canada, as follows: (1) summary of work on national income in Canada, 1918-41; (2) some principles involved in the new departure of national accounting; (3) the work of the Dominion Bureau of Statistics in preparing national accounts for Canada; (4) some illustrative applications of national accounts to economic analysis and policy; (5) problems of concepts and statistics involved in national accounting; and (6) lines of future development.

A Summary of Work on National Income in Canada, 1918-41. When it was decided in 1944 to undertake the preparation of a new set of national accounts, in line with Canadian needs and developments elsewhere, we were fortunate in having at our disposal a great deal of pioneer work on national income in Canada.

The aim of this paper is to consider the connexion between social theory and research in the field of “industrial relations,” drawing particularly upon the empirical data which have been acquired by a sociological investigation of the city of Windsor, Ontario. Even a cursory investigation of a city like Windsor indicates certain tendencies to be at work in highly unionized towns which do not seem to be adequately encompassed by the existing literature of industrial relations. We are concerned here to define what these tendencies are, to suggest some of the inadequacies of existing theory to handle them and to indicate a more adequate theoretical basis for the discussion of them.

The traditional academic and research approach to trade unionism has always been that of economics, and it is usually within an economic framework that unionism is discussed. A sociologist, however, who approaches the subject for the first time, begins immediately to wonder why trade unions should have become, in our universities and research institutions, so exclusively matter for study by economists, whereas other social groupings in which men live and to which they give their time and their loyalties—churches, for example, or gangs, or families—have not excited the slightest interest on the part of that discipline. The answer to this question has its roots, it would seem, in the classical view that since everything in the market is a commodity, and labour is in the market, therefore labour is a commodity, and as such, has to be treated and studied by the academic discipline whose job it is to analyse market factors, of which labour costs are one.

Corrupt electoral practices and controverted elections are topics which might normally be expected to go hand in hand, for the latter usually result only from the former. To some extent, the number of controverted elections is a useful criterion by which to measure the prevalence of electoral corruption; but it may be a highly misleading one, as the corruption must be discovered and proved before any election result can be affected. The criterion is therefore trustworthy only in so far as the relevant legislative devices are effective. To the degree that laws providing for the investigation of electoral malpractices are operative, bribery and personation are discouraged; weakness in these laws, on the other hand, will inevitably stimulate unscrupulous candidates and election agents to utilize any opportunities which may be turned to their advantage.

No one who has even a nodding acquaintance with the tolerant attitude taken towards political behaviour generally in the years immediately after 1867 will be surprised to hear that for some time after Confederation the whole problem of corrupt practices and controverted elections was handled in a haphazard fashion. The four provinces which came together in 1867 agreed in general terms on a definition of corrupt practice: it included the giving and receiving of bribes, whether of money, employment, or anything else; the real or threatened use of violence; treating; personation; and in some cases payment for the hauling of voters to the polls. They also had similar views as to how action for corrupt practice should be begun (by a simple petition signed by either an elector or candidate), and had adopted somewhat similar ways of trying the actual election cases, though their procedures varied in detail. Thus the Assembly of the province of Canada had a general committee of elections, appointed by the speaker subject to the sanction of the House, which was charged with preparing two panels of chairmen and members from which election committees consisting of a chairman and four members were finally chosen. Nova Scotia and New Brunswick sent controverted elections directly to a select committee. Manitoba and British Columbia, when they entered the federation, used the ordinary judicial system and tried election disputes like other cases. In the provinces where committees of the assembly were used, the procedure was similar to that of a court: the committee members were sworn to uphold justice; witnesses were examined under oath; and the parties concerned were represented by counsel. It is to be feared, however, that judicial impartiality was noticeably absent in the rendering of verdicts.

It has become an accepted fact that the mode of livelihood of a people is related in a number of ways to other parts of their culture and to some elements of their character. Anthropologists have contributed, from their relatively simple data, the clearest statements of these relationships. They have also shown that the goals and motivations in primitive economy are usually multiple and complex ones, and are as distinctive of each culture as are personality, social life, art, and religion. This demonstration has a very direct bearing on the work of those charged with planning or carrying out a programme for the welfare of a primitive people, and because of the distinctiveness of each situation, the most fully satisfactory programme can result only from an. inspection of the actual nature of the people, their culture and their resources.

The relations of the economic to the non-economic life of primitive peoples are high-lighted when a primitive group comes into contact with western culture. As a result of this contact there follows a characteristic sequence of changes in the economic life of the primitive. First of all he adopts, quite readily and often eagerly, the usually superior tools and materials, and alters some of his techniques so as to use them efficiently. The product of his labour, however, he tries to use in the same manner as before. He wants to achieve the goals which hold meaning for him, and which harmonize with his tribal system of distribution and use, its social and political organization, its religion, art, and science. He wants an axe so that he can make a canoe in a month instead of a year, a gun to hunt with or to fight the same old tribal enemies, cloth to replace the less effective barkcloth or hide.