evince a willingness and enthusiasm to contribute to the general fund of knowledge of trade diseases and mortality, they would not only perform a really valuable service to their fellow men, but they would at the same time connect their names with investigations similar to those which have contributed greatly to the fame of Ramazzini, Thackrah, Hirt, Farr, Arlidge, Oliver, and some others of the select few who have found time to give to the world the results of their inquiries into the effect of certain industries or trades upon the disease liability and mortality of those employed therein. Few, indeed, are the contributions which have been made along these lines by qualified American authorities. It may very properly be said that here is a field where the harvest is full and ripe, but the reapers few.

STANDARDIZATION OF HOUSING INVESTIGATIONS.

BY JOHN R. COMMONS, PROFESSOR OF POLITICAL ECONOMY,
UNIVERSITY OF WISCONSIN.

I had read that Glasgow was the most densely crowded of modern cities, because fourteen per cent. of the families lived in one room. After visiting one of the model tenements of the London County Council, I was asked by a Glasgow mechanic to look into his ancient rookery. The one room in which he and his family lived seemed to me to be larger than the threeroom apartment of his fellow who enjoyed the municipal socialism of modern London. The difference was that he put up his own flimsy partitions, while paternal London got the credit of relieving congestion by merely erecting permanent partitions.

In Pittsburg I was told by experts in housing investigations that the cost of housing there was greater than in any other city of the country, but when I compared the few houses that I saw with similar houses in Chicago, taking into account appurtenances, I could not see that the costs were different.

British workmen and employers contended that their lower wages were compensated by the lower cost of food as well as housing, compared with American wages, and I could not refer them to any authentic standards of food and prices, housing and rents, that would disprove their claims to their satisfaction.

If comparisons of this kind were a matter of profit and loss, standard units would long since have been devised. Such units have been worked out by the trusts, syndicates, and engineering societies, in order to bring all of their manufacturing plants, their superintendents, managers, and engineers, their inventions and experiments, to an exact comparison of efficiency based on unit costs.

There is one department of sociology which eventually will make it plain that standard units of housing, food, and occupation, are also a matter of profit and loss. This is the health, vigor, and efficiency of the working population. The trade longevity of the workman, the number of days lost through sickness, fatigue, and devitalizing, the rate of mortality, are the greatest of all matters of national business, and they are largely the results of housing, food, and occupation.

But, to what extent these different factors enter, it is impossible to say until standard units are devised by which to compare each factor with the resulting morbidity, mortality, and fatigue.

Here the problem of the economist and that of the hygienist overlap. The economist is interested in comparative cost of living, the hygienist in comparative causes of industrial efficiency. But the cost of living is really the cost of the workman's efficiency. If so, the unit of comparison which the economist wants is the same unit that the hygienist wants. Take housing as the simplest problem. The comparative cost of housing is the comparative price paid for a unit of housing accommodation. But housing accommodation is not merely floor space or "rooms per occupant": it is also location, air, ventilation, sunlight, structural condition, bath, laundry, running water, etc. These are also the conditions of health. The cost of housing is one of the costs of industrial vigor. If, then, we devise our standard unit with reference to the conditions of health, we shall have practically the standard needed for comparing prices of housing accommodation.

But the unit of housing accommodation is a complicated and elusive one. It consists of many factors, and no two individual investigators attach the same weight to each of the factors. The problem here is exactly the same as that which has been met in standardizing and grading agricultural products, such as wheat, corn, oats, butter, cheese, horses, cows, pigs, and so on. To illustrate by means of the score card used in the department of Animal Husbandry of the University of Wisconsin: A draft horse, perfect in every particular, is represented by 100 points. These are subdivided into a detailed and com

plete survey of the animal, involving 36 specifications. To each specification is given a weight, or value, of 1 to 10 points, corresponding to its importance in making up the perfect animal. This weighting is a matter of experience, and is changed from time to time. An official score card adopted by a breeders' association, or, in the case of grains, by the produce exchange, is the result of many years of experiment and improvement, both in the description of the specifications and in the weights assigned to each specification. Thus the "general appearance" of the draft horse is now given a weight of 29 points, and this is subdivided into "weight," 5 points, "form," 4 points, "quality," 6 points, "action," 10 points, and "temperament," 3 points. "Head and neck" are given 8 points, subdivided into "head," "forehead," "eyes," "ears," "muzzle," "lower jaw," and "neck," with one or two points each. These standard weights, or values, are printed in a column opposite each specification, and a second, or blank column is provided under the caption "Points Deficient." In using the score card, the "scorer" goes over the horse, noticing in detail all the points specified, and then marks down opposite each his judgment of the degree to which the animal before him is deficient in that particular point. The total of all points deficient is then deducted from 100, and the result is the grade of the animal scored. It is an interesting fact, illustrating the accuracy of this method of standardizing, that recently two horse valuers, one employed by the Wisconsin Railway Commission, and the other by the Milwaukee Street Car Company, in valuing fifty horses belonging to the company, came within one or two points of placing the same value on each horse.

In attempting to adapt the score card method to the housing problem, I have drawn up the following tentative score card for dwelling-houses. The twenty-five specifications can doubtless be greatly improved. Others might be added, and some might be dropped. The weights given to each might be changed materially. The latter, however, is not important, because, if all houses are scored by the field agents according to an agreed scale, any hygienist or economist afterwards can revise the

weighting according to his own theory of the relative weights. The main object is to agree on the specifications, and to state them in such a way that as little discretion as possible shall be left to the field agents. Where measurements are possible, this is easily done, as in the case of "window openings." Where measurements are not possible, the agents must depend on their judgment, but this judgment can be brought close to uniformity by means of "instructions for discrediting when depending on judgment." For convenience I have used only the weights 3 and 6 for those specifications depending on judgment, and have introduced the same kind of instructions as those given in the official score cards for horses and cattle. Further instructions will be found under the several specifications.

When a house is scored in this way, from the standpoint of health, we shall have the "total points deficient," and the "actual score" of that house compared with a perfect or ideal house. We are then in a position to compare the rents or cost of housing by correcting the "nominal rent" by means of the "actual score." I have suggested three standard units of comparison; viz., "rent per room," "rent per 100 sq. ft." of floor space, and "rent per 1,000 cu. ft." of air capacity. Taking "rent per 100 sq. ft.," which is probably the fairest unit under all circumstances, it can easily be seen that, of two houses renting nominally at $1.00 per month per 100 sq. ft., if the "actual score" of one is 80 and the other 50, the "real rent" of the one is $1.25 and the other $2.00 for the unit of house accommodation compared with the real rent of $1.00 for a perfect house.