ORGANIZATION OF THE STUDY

The present inquiry into the income and standard of living of unskilled laborers in Chicago had its origin in a discussion in the Committee on Family Social Work of the Chicago Council of Social Agencies over the question as to whether a proposed estimate for a revision of the Standard Budget for Dependent Families did not set a higher standard than that which could be maintained by the families of independent unskilled wage-earners. The Budget estimate,1 prepared by Miss Nesbitt for the consideration of the Committee, was more liberal in some respects than any of the previous estimates; and objection was made to it on the ground that it would provide more generously for a dependent family than an unskilled wage-earner could provide for his family. This objection revives in a different form the old question as to the applicability of what was called in England the principle of “less eligibility.” It will be remembered as one of the famous “principles” of English poor-law reform1 See Appendix A.0237in 1834 that the condition of the family living on the rates should be “less eligible” than that of the family of the self-supporting laborer.1 It is, however, both impossible and unnecessary to go into that question here.

1 See, e.g. S. and B. Webb, English Poor Law Policy, pp. 84-85, 260-263, and see Index.

The Committee, after much discussion, decided not to accept the new budget estimates until facts could be collected which would throw light on how the independent unskilled laborer and his family live today in the city of Chicago. In order to secure these facts, it was decided to ask for the co-operation of some of the large employers of unskilled labor in Chicago and of the School of Social Service Administration and the Local Community Research Committee of the University of Chicago. When this co-operation had been secured, a plan of organization2 was drawn up and accepted by all those interested in the study. According to this plan, the study was to be undertaken under the joint auspices

2 The following extract from the plan agreed upon may be of interest. “This study is to be made under the joint auspices of the Chicago Council of Social Agencies and the Committee on Local Community Research of the University of Chicago. The work of research shall be carried on under the direction of the Committee on the Study, operating as a sub-committee of the Executive Committee on Local Community Research. This sub-committee is to be composed of Prof. Harry A. Millis, Chairman, Professors Edith Abbott, Sophonisba P. Breckinridge, and E. W. Burgess, of the University of Chicago, Miss Florence Nesbitt, of the United Charities, and Mr. Wilfred S. Reynolds, Director of the Chicago Council of Social Agencies.

“The Committee on the Study shall also have charge of the assembling of material and its preparation for publication; and of all matters connected with publication, with the understanding, however, that the report shall not contain the names of the cooperating firms unless they expressly wish this to be done, nor the names of any of the persons from whom schedules are obtained. The firms shall be referred to as Company A, Company B, etc.

“The Committee will ask each firm to furnish a list of names of common or unskilled laborers who have been on the firm's payroll for the year 1924. If the records of the firm show the marital and family status of each worker, it will be asked to include in the list only those men who are married and have at least one child under fourteen years. The Committee will make this selection during the course of the study if it is not convenient or possible for the firms to do so at the beginning. When the desired information has been obtained in the schedules from approximately 500 families, the Committee will ask each firm to furnish a statement of the actual earnings for the year 1924 of each employee from whose family a schedule has been obtained. The Committee will keep a record of every name given by the firms which will show the reasons for excluding every family which was not scheduled.

“The Chicago Council of Social Agencies shall organize an Advisory Committee, to be composed of representatives of (1) the social science group at the University, (2) the industries, and (3) the representatives of the Council of Social Agencies especially interested in this study. The Advisory Committee shall be informed upon the progress of the study, and shall advise with reference to methods, emphasis and problems involved in the investigation, and shall cooperate in such other ways as opportunity may afford.”

Definition of Terms:

For the purpose of this study the following definition for common or unskilled labor was decided upon:

“Common or unskilled laborers are men who can be put to work without previous training or experience or with such instructions as can be readily given—as in a day. Those men will be excluded whose health may have been undermined by age or physical defect or who are suffering from chronic disease or from the effects of accidents. It is agreed that an effort be made to obtain a random sample of the common or unskilled labor group as above defined, with a due proportion of representation from the different wage rates paid within this group.”

It was later found necessary to modify the definition of “common or unskilled laborers” to include semi-skilled men because there seemed to be no accepted standard as to the difference between unskilled and semi-skilled labor. Another change was made necessary because some of the employers stated that their laborers needed more than a day in which to learn their job. In the end it was agreed to select men definitely not of the skilled trades who were engaged in unskilled or semi-skilled work earning approximately $30.00 a week or less. This last stipulation, in turn, could not be rigidly adhered to as some employers pay at a higher rate than others, some give bonuses which raise materially the total earnings of the men, and some have special rates of pay if the work, even of a very simple nature, is very heavy or is done under such difficult conditions as extreme heat, cold, or dampness. Within these rather elastic limits will be found the wage earners whose families were interviewed and from whom information was obtained.

0248of the Chicago Council of Social Agencies, the Local Community Research Committee and the School of Social Service of the University of Chicago; and the research was to be directed by a sub-committee of these groups. In addition to this sub-committee, the Council was to organize an advisory committee which should include not only representatives of the University and of the Council of Social Agencies, but also representatives of the industries from whose workers facts were to be obtained. This advisory committee was to pass on the plan for the study and co-operate in such ways as became necessary as the study progressed.

PLAN AND METHOD OF THE STUDY

While helping to formulate the general plan of organization, the committee in charge of the study was also working upon the plan of procedure for the investigation. It 0259was clear that there were two main subjects about which information was desired: one relating to the amount of the family income, and the other to the question of the kind of living conditions that could be provided in Chicago out of this income. Briefly stated, these two lines of inquiry were:

1. The extent to which the income of families where the chief wage-earner is an unskilled worker is augmented from sources other than the earnings of the husband and father.

2. The adequacy of the total family income derived from all sources to provide for a normal standard of living.

A collection of family budgets from wage-earners in the unskilled groups was necessary to provide data which would throw light on both of these questions. In planning a budget study, the first problem arising is whether the method of keeping accounts, obviously the most accurate and most desirable method, can be used or whether it is necessary to fall back on estimates of 02610expenditures. Since the account-book method is both expensive and slow, it was decided to use the method of estimates for as many families as could be secured from the employers’ lists, and then to select from this larger group a smaller number for actual account keeping. It was hoped that the accounts would provide a check as to the accuracy of the estimates.

With these objects in view, the investigation naturally fell into two parts:

1. A general study of the income and standard of living of an adequate number of families of unskilled wage-earners. For this study a schedule1 was prepared which was to be taken by field agents to the families whose names appeared on the lists furnished by employers. The facts to be obtained for this schedule included the family composition; the nationality of the parents and their ability to speak English; the number of persons in the household who were employed and the number of weeks in 1924 when each one was employed; the estimated earnings of each; a statement of the actual earnings of the chief wage-earners to be obtained from the employer; the amount of money contributed by each wage-earner to the family fund; the number of roomers and boarders and their contribution to the family fund. There was also a place on the schedule for an estimate by the person interviewed of the total family expenditure for 1924 for living expenses including rent, food, clothing, fuel and light, household supplies, health, education, carfare, insurance, recreation, etc.; and an estimate of savings and debts for 1924. Facts were also to be obtained about housing conditions—the size and location of the apartment or house, the number of rooms, the number of beds, possession and location of toilet or bathroom, and heating and lighting arrangements.1 See Appendix B.02711Other facts to be obtained were those which would indicate a standard somewhat above a minimum level, such as the possession of an automobile, piano, victrola, radio, or telephone; and indications of poverty as evidenced by the condition of the home and of the family. There was a place also on the schedule for an estimate of the Budget under discussion for each family and for the entry of information obtained from the Social Service Exchange showing the extent to which each family had used the free services provided by the social agencies of the community, such as dispensaries, hospitals, visiting nurses, relief agencies, or recreation centers.

2. An intensive study of a smaller number of families selected from among those giving information for the general study, with special reference to actual expenditure on living expenses. These families were to be asked to keep, for at least one month, account books in which they would enter accurately every day the exact amount of money spent on all purchases and the amount in pounds of all food purchased. It was planned from the analysis of these accounts and the comparison of these analyses with accepted standards of the food requirements for adults and children to secure information concerning the nutritive value of the diets of the families so studied, from which conclusions might be drawn as to whether the standard for food in the Chicago Budget was a reasonable one to set for a dependent family.

METHOD OF PROCEDURE

While the schedule was in course of preparation, the committee was active in enlisting the interest of representative employers of unskilled and semi-skilled laborers. When the purpose of the inquiry was explained, it was usually possible to secure a favorable response to the request for co-operation; and, all together, twelve 02812firms took part in this project. Their co-operation varied with the extent of their interest and with their individual judgment as to how much each could do. Three firms contributed both lists of names and gifts of money; one gave money alone; and eight, lists of names alone. The United Charities of Chicago and the Jewish Social Service Bureau of Chicago also contributed to the research fund. It was not possible to approach all the firms at any one time, so that the lists of names were received at intervals from January until June, 1925. The names of unskilled and semi-skilled laborers totaled 2,354.

These lists furnished by the employers were in many different forms. Some gave the name, address, occupation, and the rate of pay, and also specified whether the man was on piecework or daywork. Other employers added to these facts the total number of hours worked and a statement of total earnings for each man for the year 1924. One firm gave only the names and addresses, but divided the names into groups representing the unskilled or semi-skilled classes under such headings as the following: “Men who have worked one year or more now earning 48 cents per hour for unskilled labor, who, working a normal period, would earn $24 a week,” or “semi-skilled men who have worked one year or more at 54 cents an hour, who, working a normal week, would earn $27 per week,” or “colored janitors who have worked more than one year receiving $24 per week.” Another firm made this statement: “Our labor is in three classes: (a) common at 42½ cents an hour. There are a very few who get 43 cents to 44 cents, but these are in unpleasant jobs (you will see that they are few because our common-labor rate with a minimum of 42½ cents averages under 43 cents); (b) semi-skilled, 44½ cents to 52½ cents; (c) skilled, 53 cents to 82 cents.”

02913

The list from this firm included men from the first two classes. Two other firms gave the marital state of the men and the total number of children and number of dependent children. One firm, in addition to names and addresses, gave average earnings per week for each man in 1924 and 1925. Another gave the names and addresses classified into wage groups—as those earning between $1,400 and $1,500, $1,500 and $1,600, etc., for the year 1924.

When each list was received, a card catalogue was made with the name and address of each man on a separate card. These names were then assigned to the field agents, each of whom worked in different sections of the city. This work began on February 20, 1925. The agents then went from one address to another, interviewing the housewife in each family and asking for the information necessary for the schedules. When the schedules were completed, the agents turned them in to the office together with a report on every family from whom a schedule had not been obtained. There were 1,915 families from whom no information could be secured, but a total of 439 schedules were obtained from families on the lists, and 37 more from families upon whom the agents chanced to call in their search for certain addresses. Nine of the latter group had to be discarded as they were found to be doing skilled work. In many cases those who gave the information were greatly interested in the study and answered the questions with great willingness. If any hesitation or suspicion were shown, the agent was instructed to leave the house without exerting undue pressure in an effort to fill the schedule.

No information was secured from 1,915 families for the following reasons:

1. In 625 cases an incorrect address was given. This was probably due to different causes, such as the inability 03014of the men to speak English well enough to be understood when giving their address at the employment office of the plant; the desire of some of them to withhold this information from their employers; the difficulty on the part of the employment offices in keeping such records up-to-date; and the difficulty the field agents found in pronouncing foreign names and in making themselves understood, especially when the women did not speak English.

2. In 715 instances the names given did not comply with the requirement that the man should be married and have at least one dependent child in the home. Of these, 620 were men who had either no children or children who were grown up and self-supporting; 38 were single men; in 19 instances the man had a wife and family in Europe; 38 were married but had no homes because of the death or permanent incapacity of their wives, or because of separation or divorce.

3. In 130 cases the housewife refused to give the information.

4. In 260 cases no information could be secured on the first visit either because of the language difficulty or because it was not possible to find anyone at home. At least one subsequent visit was made in every case but no contact was ever established with these families.

5. In 33 cases the families were living outside of the city limits and so were not eligible for the study.

6. In 9 cases the families had left town since 1924.

7. In 7 cases the living arrangements were so complicated that, in spite of the willingness of the families to give the information, schedules could not be obtained.

8. In 42 cases the man was not regularly employed in 1924.

9. In 94 instances the families were not visited as it 03115was necessary because of the limitations of time and money to stop the field work on August 31, 1925.

When the schedules were completed, a letter was sent to each of the co-operating firms asking for a statement of actual earnings for each man for the year 1924. The replies to this request were as follows:

FirmNames SuppliedSchedules ObtainedStatements of EarningsA623183173J53010093C2582828D2501414K1692323E10988G1053932F1001616I8632L802019B4455M*37†13‡Total2,345476426* This is the group whose names were not on lists furnished by employers.† Nine of these are skilled workmen and are not included in the tabulations.‡ Three of these are skilled and are not included in the study.

Firm A could not give yearly estimates for 10 men because through a clerical error these men were on the lists although they had been employed only for a short time in 1924. The same thing is true in the case of firms G, I, and J. In group M it was impossible to identify the firms for whom 9 men were said to work, so on information could be obtained as to their wages. Letters were written to the 24 firms which could be identified, asking information concerning the wages of the other 28 men. One of these firms—a public department—refused information; 4 could not find the names on their pay-roll for 1924; 14 sent the desired information; and 5 firms failed to reply. They all showed an interest in the study 03216and a desire to co-operate. Several requested that the information be considered confidential.

The field work occupied approximately eight months. The schedules were corrected immediately after they were handed in, and the work was so organized that the tabulation of the schedules was begun in October. In the material which has been assembled, facts should be found which will throw new light on two important questions (1) the total income of families of unskilled laborers; together with information showing how far this income is derived from the chief wage-earner, to what extent it is supplemented from other sources, what those sources are, and whether, when all possible sources have been utilized, a “normal standard” of living is maintained; (2) the relation of this standard to that set up by the Chicago Standard Budget in order to ascertain whether the Chicago Budget is seeking to create for dependent families an unreasonably high standard of living. In the following chapters the results, as finally analyzed, are presented.

AGES OF CHIEF WAGE-EARNERS ANDLENGTH OF EMPLOYMENT

Before discussing facts relating to employment and wages, there are certain data to present about this group of men which are of significance and interest. These deal with age, nationality, and race. Table I gives the age groups to which the men belong.

This table shows that 333, or approximately three-fourths of the men, were young men, under forty-five years of age. In considering the fact that 112 men, or 25 per cent, were forty-five years of age or over, it must be borne in mind that this group of unskilled laborers belonged in a very fortunate group, as they were men whose names were carried on the books1 of their employers1 This does not necessarily mean that these men were at work continuously without any days of unemployment in 1924.03418during the entire year. It is safe to assume that, in general, this would indicate a group of men who had been a relatively long time on the employers’ pay-rolls, who were probably in the better class of employees, and who would be among the last to be laid off in time of industrial depression. This assumption is borne out by the facts in Tables II and IIA in regard to the period of employment with the different firms.

Age GroupsNumberPer Cent DistributionTotal467Not reported22Total reported455100.0Under 25 years51.125 and under 35 years10824.335 and under 45 years22049.445 and under 55 years9320.955 and under 65 years173.865 years and over2.5

TABLE IWage-Earners of Specified Ages; Number andPer Cent Distribution

While very often the unskilled or semi-skilled laborer is the casual laborer, going from job to job, and easily replaced because of the simple nature of his work, these men had, on the whole, been a long time on the pay-rolls of the firms employing them. Only 49, or 10.8 per cent, had been employed for less than two years. Of even greater importance is the fact that 47.8 per cent had been in the employ of one firm for at least five years, and 23 per cent had a period of employment of ten years or over. In the light of these figures, the assumption that these men form a very favorably situated group of unskilled or semi-skilled workers whose work records were good and whose services were valuable to the firms, seems to be sustained.

03519Chief Wage-EarnersNumber of Years Employed by FirmNumberPer Cent DistributionTotal467Not reported15Total reported452100.0Less than one15*3.31 and less than 2347.52 and less than 38418.63 and less than 47416.44 and less than 5296.45 and less than 1011224.810 and less than 155111.315 and less than 20316.920 and less than 251613.525 or over61.3

TABLE IIChief Wage-Earners Classified According to Lengthof Employment by Firm; Number andPer Cent Distribution

* These men were employed less than one year by the firms supplying their names. They were, however, according to their own statements, fully employed during 1924 and so were included in the study.

TABLE II AChief Wage-Earner Classified Accordingto Length of Employment by Firm;Cumulative Percentages

Number of Years Employed by FirmCumulative Percentage

Total100.0

Less than 13.3

1 and over96.7

2 and over89.2

3 and over70.6

4 and over54.2

5 and over47.8

10 and over23.0

15 and over11.7

20 and over4.8

25 and over1.3

RACE AND NATIONALITY OF CHIEF WAGE-EARNERS

Since it was felt that income was the determining factor in the present study, no effort was made to secure from the employers any specified proportion of the different races and nationalities. The resulting numbers of

Chief Wage-EarnersRaceNumberPer Cent DistributionTotal467100.0White38081.4Colored8718.6

TABLE IIIRace of Chief Wage-Earners

RaceAdult Male Population of Chicago in 1920*Chief Wage-EarnersNumberPer Cent DistributionNumberPer Cent DistributionTotal874,239100.0467100.0White828,95194.838081.4Negro42,8374.98718.6All others2,451.3

TABLE IVRace of the Adult Male Population of Chicago Twenty-One Yearsof Age and Over, in 1920, and of the Chief Wage-EarnerStudied; Number and Per Cent Distribution

* Fourteenth Census of the United States, Vol. III.

white and colored families are shown in Table III. This table shows 380 white families and 87 Negro families. To give this statement more significance, a comparison was made with the figures for the adult male population in Chicago in 1920. The result is shown in Table IV.

When the relative numbers of Negro and white workers are compared with the Census figures for 1920 03721(Table IV), it will be seen that, while 18.6 per cent of the unskilled laborers studied were Negro, only 4.9 per cent of the adult male population in Chicago in 1920 were of that race. The difference is undoubtedly explained by the fact that, especially since the war and the restriction of immigration, the Negro has been taking the place of

Country of BirthWage-EarnersNumberPer Cent DistributionTotal647Not reported3*Total reported464100.0United States†13128.3Foreign born:Poland13829.7Italy7215.5Other Slavic countries‡326.9Lithuania286.1Czechoslovakia91.9Germany91.9Hungary91.9Holland71.5Scandinavia51.1Ireland51.1All others194.1

TABLE VCountry of Birth of Chief Wage-Earners; Numberand Per Cent Distribution

* These three were foreign-born, but country of birth was not specified.† Includes 87 Negroes.‡ Includes Russia and Jugoslavia.

the Southeastern European immigrant in doing the simpler kinds of labor.

Table V shows the country of birth of the chief wage-earners. Among them there is a great diversity of nationalities, with nearly three-fourths of the whole number foreign-born. This is to be expected, because in the past the ranks of unskilled labor have been so largely recruited from the immigrant peoples from Europe, 03822while the native-born tend to go into the skilled trades or into some other of the higher occupational callings. From this table it can be seen that 131, or 28.3 per cent, were native-born, but it should be noted that 87 of these were Negroes, leaving only 44 native 44 native-born white men in this group. Of the foreign-born, the largest groups are the Poles, the Italians, and the Lithuanians. This is not surprising, as the unskilled laborer most commonly comes from Southern or Eastern Europe and is a part of the so-called “new immigration.” When to

NativityAdult Male Population of Chicago in 1920*Chief Wage-EarnersNumberPer Cent DistributionNumberPer Cent DistributionTotal874,239100.0467100.0Native-born469,82353.713128.1Foreign-born404,41646.333671.9

TABLE VINativity of Adult Male Population of Chicago in 1926 and ofChief Wage-Earners Studied; Number andPer Cent Distribution

* Fourteenth Census of the United States Vol. III

these three nationalities are added the Hungarians, the Czechs, the Russians, and the Jugoslavs, the result is a total of 288, or 62 per cent from Southern and Eastern Europe; while the Northern and Western Europeans number 26, or 5.6 per cent. A comparison of the nativity of these wage-earners with that of the adult male population of Chicago in 1920 is shown in Table VI.

A comparison of this nationality grouping with figures from the 1920 Census for the adult male population, shows that as regards the distribution of native and foreign-born the figures are very different. In Chicago in 1920, 53.7 per cent of the adult male population were 03923native-born, and 48.3 per cent foreign-born. In the group under consideration, 28.1 per cent were native-born, and 71.9 per cent foreign-born. This is due, of course, to the fact that this group is composed only of unskilled and semi-skilled laborers and belongs in the lower economic level where the foreign-born predominate.

ABILITY TO SPEAK ENGLISH

While there were 336 foreign-born laborers studied, it is interesting to note that only 79 of them, or 23.5 per cent, were unable to speak English. As it is shown later in this report that almost half of them had been in the employ of their firms for five or more years, and probably in this country for even a longer period, this is not a surprising proportion. It must also be remembered that there was undoubtedly a larger proportion of non-English-speaking husbands among the families from whom schedules could not be secured (see page 14).

EARNINGS OF CHIEF WAGE-EARNERS

Passing on to the important subject to wages, it is interesting to note that the range of wages paid was very wide, varying from $800 a year to $2,200 and over. The reason for this, as has already been stated, is that there seems to be no exact definition of the terms “unskilled” and “semi-skilled” labor, and wages vary greatly between firms and are influenced by the conditions of work. Higher pay frequently accompanies work which, while simple in itself, is done under difficult conditions. Besides this, one of the firms concerned had a bonus system which brought the total payment to each man to an unusually high figure. Schedules were obtained for only 20 men from this firm, 13 of whom received over $1,500, and 6 of these received over $1,800.

It must be recalled here that data regarding wages 04024were obtained from two sources: (1) estimates made by the housewives, at the time the schedule was taken, of their husbands’ yearly earnings; (2) statements obtained directly from the employers showing the exact yearly earnings of each man. In all the tables concerning the

Earnings of Chief Wage-EarnerChief Wage-Earners of Specified RaceTotalWhiteColoredNumberPer Cent DistributionNumberPer Cent DistributionNumberPer Cent DistributionTotal46738087Not reported44*377Total reported423100.0343100.080100.0$800-$89961.451.511.3$900-$999112.672.045.0$1,000-$1,099409.5226.41822.5$1,100-$1,1995011.8339.61721.2$1,200-$1,2996314.94814.01518.7$1,300-$1,3996314.96017.533.8$1,400-$1,4998119.27421.678.7$1,500-$1,5994510.63710.8810.0$1,600-$1,699245.7226.422.5$1,700-$1,799204.7154.356.3$1,800-$1,89961.461.8$1,900-$1,99971.772.0$2,000-$2,0992.52.6$2,100-$2,1994.941.2$2,200-$2,299$2,300-$2,3991.21.3

TABLE VIIEarnings of Chief Wage-Earners Classified by Race; Numberand Per Cent Distribution (Data from Pay-Bolls)

* It was impossible to secure the employer's statements of earnings for these man, most of whom were not on the lists furnished for the study. The statements of estimated wages were secured for each man, but were not used in this table.

earnings of the chief wage-earner, the data presented are those furnished from the pay-rolls of the employers. Tables VII and VII A Show the range of earnings of the 467 chief wage-earners.

A study of these tables shows that for the great majority of the men, wages ranged between $1,000 and 04125$1,800. Only 17 men, or 4.0 per cent, received less than $1,000; and only 20 men, or 4.7 per cent, earned $1,800 or more. It is probably more important to note the numbers which fall within the different wage groups, as this illustrates the earning power of the chief wage-earners in the 467 families. It will be seen that 107 men, or 25.3

Earnings of Chief Wage-EarnerChief Wage-Earners of Specified RaceCumulative PercentagesTotalWhiteColoredLess than $9001.41.51.3Less than $1,0004.03.56.3Less than $1,10013.59.928.8Less than $1,20025.319.550.0Less than $1,30040.233.568.7Less than $1,40055.151.072.5Less than $1,50074.372.681.2Less than $1,60084.983.491.2Less than $1,70090.689.893.7Less than $1,80095.394.1100.0Less than $1,90096.795.9Less than $2,00098.497.9Less than $2,10098.998.5Less than $2,20099.899.7Less than $2,30099.899.7Less than $2,400100.0100.0

TABLE VII AEarnings of Chief Wage-Earners Classified by Race;Cumulative Percentages

per cent, earned less than $1,200; 233, or 55.1 per cent, earned less than $1,400; 359, or 84.9 per cent, earned less than $1,600; while 403, or 95.3 per cent, earned less than $1,800.

In comparing the wages of the white workers with those of the colored workers, a great difference is apparent. While 50 per cent of the Negroes earned less than $1,200, only 19.5 per cent of the white workers were in this group. Nearly three-fourths of the Negroes 04226earned less than $1,400, as compared with 51.0 per cent of the white workers. In the wage groups earning over $1,500 there is closer relationship in the percentage of both races; but, while there are no Negroes earning over $1,800 there are twenty white workers—a small number, but one further indication of the disparity existing between the earnings of the two races.

EARNING OF MEN IN RELATION TO AGE GROUPSAND LENGTH OF TIME WITH FIRMS

After analyzing the earnings, it is interesting to consider whether there is any relation to be found between

Earnings of Chief Wage-EarnerNumber of Chief Wage-Earners of Specified AgesTotalUnder 2525 to 3435 to 4445 to 5455 to 6465 or OverNot ReportedTotal46751082209317222Not reported4417189Total reported4235912028417222$800-$8996411$900-$9991112521$1,000-$1,09940113146213$1,100-$1,1995011201135$1,200-$1,29963116291313$1,300-$1,39963935142121,400-$1,49981118421523$1,500-$1,599451112571$1,600-$1,69924111921$1,700-$1,79920610211$1,800-$1,899624$1,900-$1,99972311$2,000-$2,099211$2,100-$2,1994211$2,200-$2,299$2,300-$2,39911

TABLE VIIEarnings of Chief Wage-Earners during 1924 Classified by Age(Data from Pay-Rolls)

the earnings of these wage-earners and their ages. Table VII indicates that such a relation does not exist to any 04327marked extent. The younger men do not appear to any disproportionate extent in the higher wage levels. It is only in the age group between fifty-five and sixty-four that there seems to be a larger proportion in the lower levels, although this group also has its representation in the higher wage groups.

Earnings of Chief Wage-EarnerNumber of Chief Wage-Earners Employed by Firm for a Specified Number of YearsTotalLess Than 11 and Less Than 22 and Less Than 33 and Less Than 44 and Less Than 55 and Less Than 1010 and Less Than 1515 and Less than 2020 and Less Than 2525 or OverNot ReportedTotal4671534847429112513116615Not reported4411472281225Total reported423430777227104502914610$800-$899611121$900-$99911121111211$1,000-$1,099402113856122$1,100-$1,19950571041092111$1,200-$1,299631710103159422$1,300-$1,399631513127126223$1,400-$1,499815161262365512$1,500-$1,59945167117841$1,600-$1,699242475312$1,700-$1,799201129421$1,800-$1,899611211$1,900-$1,9997111121$2,000-$2,099211$2,100-$2,199412$2,200-$2,2991$2,300-$2,39911

TABLE IXEarnings of Chief Wage-Earners Classified by Number of Yearswith Firm (Data from Pay-Rolls)

A question closely related to this one is the length of service with the firm and its possible relation to wages. This is shown in Table IX.

On the whole, these men had been in the employ of their firms for unusually long periods for this type of labor, as 47.8 per cent had a period of service of five 04428years or more and 23 per cent had a period of service of ten years or more. The table, however, fails to show that higher wages accompany long periods of service; but, on the other hand, it indicates that in unskilled and semi-skilled labor the pay is fixed by the kind of work done and not by length of service.

Earnings of Chief Wage-EarnerNumber of Chief Wage-Earners Employed by Firm a Specified Number of YearsTotalLess Than 11 and Less Than 22 and Less Than 33 and Less Than 44 and Less Than 55 and Less Than 1010 and Less Than 1515 and Less Than 2020 and Less Than 2525 or OverNot ReportedTotal380132457552293513016613Not reported3710352251225Total reported3433215253208850281468$800-$899511111$900-$9997111211$1,000-$1,09922116424121$1,100-$1,199332134992111$1,200-$1,2994814761129422$1,300-1,399601511127116223$1,400-$1,499744131152265512$1,500-$1,5993716612831$1,600-$1,699222275312$1,700-$1,799151842$1,800-$1,899611211$1,900-$1,9997111121$2,000-$2,099211$2,100-$2,1994112$2,200-$2,39911

TABLE XEarnings of Chief Wage-Earners Classified by Number ofYears with Firm: White Workers Only

A separate study of the same subject was made for the different races. The results for the two groups are shown in Tables X and XI.

In studying these two tables, the striking fact which appears at once is that it is almost entirely among the white laborers that the very long period of service, ten years or more, appears. There are 103 white laborers in 04529these groups, as contrasted with one Negro. This, of course, is at least partially explained by the fact that the great migration of Negroes to Chicago has occurred during the last ten years and that it is within this period that they have begun in considerable numbers to take the places hitherto held by the unskilled white laborers.

Earnings of Chief Wage-EarnerNumber of Chief Wage-Earners Employed by Firm a Specified Number of YearsTotalLess Than 11 and Less Than 22 and Less Than 33 and Less Than 44 and Less Than 55 and Less Than 1010 and Less Than 1515 and Less Than 2020 and Less Than 2525 or OverNot ReportedTotal87210271971912Not reported71123Total reported8019251971612$800-$89911$900-$99941111$1,000-$1,09918174321$1,100-$1,199173671$1,200-$1,2991533423$1,300-$1,399321$1,400-$1,499713111$1,500-$1,59981151$1,600-$1,69922$1,700-$1,799511111

TABLE XIEarnings of Chief Wage-Earners Classified by Number ofYears with Firm: Colored Workers Only

These tables show also that in the separate racial groups there is again no relationship between wages and length of service. The white laborers, as it has already been shown, earn higher wages than the Negroes; but in neither case do their wages rise with length of service.

TIME LOST

While this study is concerned only with men who were fully employed, an attempt was made to secure information as to the amount of time that each man lost 04630during the year and the reason for this lost time. The statements obtained, however, must be considered only as estimates as they were given by the housewife or

Causes of Time LostChief Wage-EarnersNumberPer Cent DistributionTotal467Not reported6Total reported461100.0No time lost22248.2Sickness13829.9No work7616.5Sickness and no work255.4

TABLE XIICauses of Time Lost by Chief Wage-Earners (Datafrom Family Schedules)

Number of Weeks IdleChief Wage-EarnersNumberPer Cent DistributionTotal467Not reported12Total reported455100.0None22248.8Less than 24610.12 and less than 46113.44 and less than 6347.56 and less than 8214.68 and less than 12398.612 and less than 16214.616 and less than 2061.320 or more51.1

TABLE XIIILength of Time Lost by Chief Wage-Earners (Datafrom Family Schedules)

some other member of the family at the time the schedule was taken; they have not been verified from the employers’ records. The first questions to be considered in 04731this connection were the number who were idle for some length of time and the reasons therefor. This is shown in Table XII.

For only 222 men, or 48.2 per cent, was no idleness reported during the year. Of the 239 reporting idleness,

Earnings of Chief Wage-EarnerNumber of Chief Wage-Earners Who Were Idle a Specified Number of WeeksTotalNot IdleLess Than 22 and Less Than 44 and Less Than 66 and Less Than 88 and Less Than 1212 and Less Than 1616 and Less Than 2020 or MoreNot ReportedTotal4672224661342139216512Not reported4414162515451Total reported423208455532163816211$800-$89963111$900-$9991162111$1,000-$1,099401498112113$1,100-$1,1995029764121$1,200-$1,29963295839414$1,300-$1,3996335785242$1,400-$1,499813731268933$1,500-$1,59945254432421$1,600-$1,699241033512$1,700-$1,79920824321$1,800-$1,899651$1,900-$1,99973211$2,000-$2,099211$2,100-$2,199422$2,200 and over11

TABLE XIVNumber of Weeks Idle and Earnings of Chief Wage-Earners(Data for Earnings from Pay-Rolls; Data for TimeLost from Family Schedules)

138 gave sickness as the cause, while 76 reported “no work.”

The amount of time lost was the next subject to be considered and the data on this point are shown in Table XII. These periods of idleness are, on the whole, rather long, as only 107 men report lost time for 4 weeks or less, 04832while 126 report longer periods of idleness, with 11 of these reporting 16 weeks or more with no work.

An attempt was made to see if there was a relationship between the amount of time lost and wages, and these facts are presented in Table XIV.

Table XIV does not indicate that the men with the longest periods of idleness were in the lowest wage groups. In fact, some of the men for whom long periods of lack of work were reported are in the high wage groups. The reason for this may be that they were in the higher-grade occupations, or it may be merely due to the inaccuracy of the statements obtained on the schedules from the housewives. It is often difficult for the wife to remember exactly the length of an illness or the month, or even the year, in which it occurred; and frequently the wife does not know definitely the length of a layoff. Moreover, in families of this economic level the situation to be faced when the main source of income is withdrawn, is so serious that there might well be a tendency to exaggerate its length.

COMPARISON OF EARNINGS AS ESTIMATED AND ASREPORTED BY THE EMPLOYERS

When this study was planned it was felt that, because of the co-operation of the employers and their willingness to furnish reports of the total earnings of each man for the year 1924, it would be possible to secure facts which might throw light on the validity of the methods which have been employed in other studies to obtain information concerning earnings. Generally speaking, it may be said that the method of estimating has been used most frequently. That is to say, when a family is visited, the housewife is asked for a statement of the earnings of the wage-earners in the family. This 04933was the method used in Dr. Chapin's study,1 and he comments upon it as follows:Among the difficulties encountered was a tendency in some cases to exaggerate in statements of earnings and expenditures, partly from pardonable pride in making as good a showing as possible, partly from a method of estimating food expenditures on the basis of a week's outlay. ... The opposite tendency to underestimate appeared in a few cases, prompted apparently by a desire to make out as bad a case as possible....

1 Robert Cait Chapin, Standard of Living Among Workingmen's Families in New York City (New York: Russell Sage Foundation, 1909), pp. 30-31.

It would be absurd to claim for these family reports the exactness of a bank statement. But, although only approximations, they were carefully prepared. ... No doubt the total income was often misstated through failure to make an exact allowances for days of unemployment. ... Some of these inaccuracies would counter-balance one another, and be eliminated in the process of averaging. Thus, the tendency to exaggeration would be offset by the tendency to concealment. It would not be safe to suppose that all errors would disappear by any such magic process, however, especially with no more cases than in the present instance.

In the studies of the cost of living made by the Bureau of Labor Statistics of the United States Department of Labor, the method used is described as follows:The data for this study were secured, as in former studies, through personal visits by agents of the Bureau to the homes of wage earners and small salaried men, where, by means of interviews with the wife or other members of the family, information was obtained relative to the income and expenditures of the family for the period of one year.2

2 Cost of Living in the United States, 1924.

Many of the studies conducted by the United States Children's Bureau, in dealing with problems of child welfare, have laid stress on the financial condition of the families, studied, and for this reason have attempted to secure information concerning earnings as the most important 05034source of income. Here again, the usual method1 has been to question the housewife as to the earnings of the employed persons in the family. While no claim is made that these estimates are exact, the attitude of the Bureau is expressed in the following sentence.2 “It is believed the information secured is sufficiently accurate for the purposes for which it has been used.”

1 U.S. Children's Bureau Publication No. 112, Infant Mortality: Results of a Field Study in Gary, Indians (1929): ibid., No. 102, Helen Russell Wright. Children of Wage-Earning Mothers (1922).2 Ibid., No. 106, Child Labor and the Welfare of Children in an Anthracite Coal Mining District.

In one inquiry made under the direction of the Woman's Bureau,3 a method of securing earnings was used which is in some respects similar to the one used in this study. It is described in the report as follows:In making the investigation the method followed was to interview all men and women in each plant, securing from them information as to their industrial history and present condition: i.e., present wage and occupation, age at beginning work, time in the trade and in present occupation, number of weeks not at work during the past year, reasons for not working, days worked during the past week, usual daily and weekly hours. ... When this information had been secured for all the men and women employed in the plant, the pay-rolls were studied and the weekly earnings of each person for as many weeks as he or she had been with the firm during the year April, 1919, to April, 1920, were recorded with the rest of the information secured about him.

3 U.S. Women's Bureau Bulletin No. 30. The Share of Wage-Earning Women in Family Support.

In this way both estimated and actual earnings were secured, but the figures taken from the pay-rolls were used in drawing up the report and no effort seems to have been made to compare the accuracy of the estimated earnings with the actual earnings.4

4 Reference might also be made to another method of using employers’ pay-rolls as described by B. Seebohm Rowntree in his well-known book Poverty, a Study in Town Life (London, 1901).05135

In this study, as has been stated, the employers have provided the actual earnings for the year 1924 in almost every case in which a schedule was obtained. As these statements, for the most part, were not obtained until after the family had been visited and the information secured, the investigators were able to get from the families their estimates of the wages uninfluenced by any knowledge of what the actual earnings were. The method of obtaining these estimates was usually that

Estimated Earnings in Excess of, or Below, Actual EarningsWage-EarnersNumberPer Cent DistributionTotal467Not reported46*Total reported421100.0Estimated earnings below actual earnings28768.2Estimated earnings in excess of actual earnings12128.7Estimated earnings the same as actual earnings133.1

TABLE XVComparison of Estimated Earnings with Actual Earnings ofWage-Earners

* In 44 of these cases the employers’ statement of actual earnings was not reported; in the other 2 cases the estimated earnings were not reported.

of asking the housewife what her husband earned every week, and by multiplying that figure by the number of weeks during the year that he was reported to have worked at that wage. A comparison of the earnings as thus estimated and as reported by the employers is given in Table XV.

When a comparison was made between the estimates and the employers’ statements, it was found that only thirteen estimates were entirely correct and that these estimates were either for men who were paid a certain fixed sum by the month, or that these families had been given statements of earnings by the employers, so that in reality their statements were not estimates, but the 05236actual earnings for 1924. In 287 instances, or 68.2 per cent of the cases, the estimated earnings were smaller than the actual earnings; and in 121, or 28.7 per cent, the estimated earnings were larger. The percentages of error in the estimates are shown in Table XVI.

Percentage of Error in Estimated EarningsWage-Earners Whose Estimated Earnings Are Above or Below Actual EarningsTotalAboveBelowNumberPer Cent DistributionNo ChangeNumberPer Cent DistributionNumberPer Cent DistributionEarnings Not ReportedTotal4671312128746Not reported4646Total reported421100.013121100.0287100.0No difference133.113Less than 510524.95041.35519.25-98319.72722.35619.510-1910023.83024.87024.420-295312.6108.34315.030-39276.443.3238.040-49112.6113.850-5992.193.160-6971.772.470-792.52.780-891.21.490-992.52.7100-14951.251.7150 and over3.731.1

Table XVIPercentage of Error in Estimated Earnings as Compared withActual Earnings Reported by Employers

As to the extent to which there was an overestimate or an underestimate, Table XVI shows that of the men whose estimated earnings were smaller than the actual earnings (whose wives were therefore underestimating), there were only 55, or 19.2 per cent, of the total number in this group, with an error of less than 5 per cent; while in the group of 121 who had overestimated their earnings 05337there were 50, or 41.3 per cent of the total, with the same percentage of error. If there is added to this group those having a percentage of error of from 5 to 10 per cent, the results show only 111, or 38.7 per cent, of those underestimating, and 77 or 63.6 per cent of those over-estimating within these limits. Among those with a very large percentage of error—20 per cent and above—the opposite is true, that is, those who underestimated show a larger proportion, 106, or 36.9 per cent; while the over-estimates numbered only 14, or 11.6 per cent of that group. The differences between estimates and actual earnings in a few cases were so large that the estimate would have to be more than doubled to equal the actual earnings.

From these figures it seems apparent that there was a strong tendency to underestimate earnings. While it is impossible to give a definite reason for this, there are several possible explanations. In the first place, the man may not give his entire earnings to his wife, but may simply give her the amount he thinks she should spend for the family. Then again, there might easily be a feeling among people of this economic level that if their circumstances should appear too comfortable, steps might be taken to lower wages on the ground that they had more than they actually needed. A third explanation may be in the fact that several times the investigators were asked if they came “from the government” and had anything to do with the income tax.

Further evidence of the unreliability of these estitimates is shown from a consideration of the group as a whole without regard to the question of whether the error arises through overstatement or understatement. Even if a difference of less than 10 per cent is conceded to be so small as to be of very little importance, there still remain 220 families, or 52.3 per cent, whose estimates 05438show larger errors. Moreover, 120 families, or 28.5 per cent of the whole group, made estimates with errors of 20 per cent or more.

All these facts seem clearly to indicate that, at least for a group of the kind studied, it is not safe to rely on estimates alone in obtaining facts concerning wages. This tendency—to underestimate, and to do so with a substantial divergence from the facts—is so pronounced that it furnishes strong evidence that estimates of wages and earnings, such as have been frequently accepted in other inquiries as reasonably accurate, are far from reliable. Moreover, attention should again be called to the fact that the investigators who secured these estimates were all persons who had had experience in family case work and who know something of life among families in the low-wage groups. Therefore the inaccuracies in the estimates cannot be charged to lack of intelligence or experience among the investigators, but are apparrently largely inherent in the method itself.

DEPENDENT CHILDREN

Before discussing the number of dependent children in the families, it is necessary to define the word “dependent” as used in this connection. For the purposes of this study, a dependent child is one who is dependent on the family fund for support. No age limit has been set, and 40 children of sixteen or over were included who were either attending school or were ill, and therefore were contributing nothing to the family fund. On the other hand, there have also been included, as dependent, 35 children of thirteen, fourteen, and fifteen years of age who have worked after school or during the summer holidays at such jobs as selling newspaper, peddling, delivering parcels, picking berries, and caddying at country clubs. These children contributed small amounts to the family fund; but as they were all attending school and were largely supported from the earnings of others, it seemed fairer to consider them as dependent children rather than as working children.

The number of persons in the “standard” or “normal” family, which is supposed to consist of the father and mother and three dependent children, has been discussed for some years in relation to the “living wage” and more recently in connection with family-allowance schemes. Mr. Rowntree in his book The Human Needs

05741Number of Persons in HouseholdWage-Earners’ HouseholdsNumberPer Cent DistributionTotal467Not reported†2Total reported465100.03469.947616.458418.167415.976514.085812.59296.210153.21171.51251.1133.614 or more3.6

TABLE XVIINumber of Persons in Wage-Earners’ Households;*Number and Per Cent Distribution

* The household includes all members of the family, dependent relatives, and boarders and † It was impossible to leave the exact number of boarders and roomers in these two Number of Persons in FamilyWage-Earners’ FamiliesNumberPer Cent DistributionTotal467100.035912.649620.559720.866814.675511.884910.59265.610122.6112.412 or more3.6

TABLE XVIIINumber of Persons in Wage-Earners’ Families;*Number and Per Cent Distribution

* The family includes the father, mother, all children, and all dependent relatives.05842

of Labor1 studied the census figures for 2,161 families in York, England, which he considered “completed families”—that is, where the mother was between forty and forty-five years of age. The purpose of the study was, as he expresses it, “to guide us in deciding what number of dependents should be allowed for in fixing the minimum wage to be paid to men.” His conclusions, drawn from a considerable body of statistical material, were that in approximately fifty per cent of all families there are at some time or other three or more dependent children to be supported by the chief wage-earner. Hence, a minimum wage should be based on the needs of families of this size.

1 B. Seebohm Rowntree, The Human Needs of Labor (London, 1918).

In a discussion of the minimum wage and the schemes for family allowances in Australia, Miss Rathbone2 comments upon the fact that the five-member family had been accepted as standard by all arbitration and wage boards in that country, with the exception of the Board of New South Wales which has the four-member family as standard. In 1920 a Royal Commission on the Basic Wage was appointed in Australia which she thinks “represents the first attempt ever made.... by the government of any nation to put a real content into the empty phrase ‘a living wage’ based on the fiction of the uniform typical family.” after a careful investigation, this commission in its report gave out the following figures on the family composition of wage-earners:27.0 per cent are bachelors or widowers without children24.7 per cent are married couples without children or with nodependent children under fourteen16.6 per cent have one dependent child18.0 per cent have two dependent children8.8 per cent have three dependent children9.9 per cent have more than three dependent children

2 Eleanor F. Rathbone, The Disinherited Family(London, 1924).05943

In the light of these figures, Miss Rathbone maintained thatat present the industries of the Commonwealth pay as if the children of the Commonwealth were 3,000,000 (i.e. three children for each 1,000,000 employees). In point of fact, the children of employees in the Commonwealth number 900,000. ... There is little doubt that the present quasi-submergence of employees with families is due to ignoring the true incidence of the actual cost of living. From the produced wealth of the country its children have less than enough in order that the unmarried childless may have more than enough.

Because of this, she urges the adoption of a minimum wage large enough for every worker to support two people, with an additional allowance for dependent children to be paid either by the state or from occupational pools organized by different industries.

In a more recent book,1 in which a strong plea is made for the family allowance instead of the minimum wage, Professor Douglas, after studying the United States Census figures, the material taken from the report of the Health Insurance Commission of the State of Illinois, and the report of the Untied States Coal Commission, comes to the conclusion thattaking the country as a whole, the adult male workers on the average do not have four additional persons dependent on them. ... There were, indeed, in 1920, 33,600,000 children under fifteen years of age, and 32,200,000 males of twenty years and over. This amounted to only 1.04 children for every adult male. Even if the full burden of support fell upon the 28,200,000 occupied males, there would have been only 1.19 children for every such male worker..... At the most, the total number of dependents to be supported by each adult male worker could not amount to more than 2.35 persons and not 4.0.

1 Paul H. Douglas, Wages and the Family(Chicago, 1925).

The interest in this subject and the attempts to throw light upon it confirm the following statement made by Professor Bowley:Very little is known accurately about one of the subjects which one might have expected to have found among the earliest of statistical 06044investigations, viz., the composition of ordinary families with respect to sex, to age, and to the number of earners and dependents, and the relative number of families of various compositions.1

1 A. L. Bowley. “Earners and Dependents in English Towns in 1911,” Economics, May, 1921.

In view of this discussion, the facts obtained in this inquiry relating to families of different sizes in the group studied are interesting. The data on this subject are shown in Tables XIX and XIX A. Since there were included in the study only those families having at least

Number of Dependent ChildrenWage-Earners’ FamiliesNumberPer Cent DistributionTotal467100.0None*2.417716.5211224.0310322.047916.95418.86347.37 or more194.1

TABLE XIXNumber of Dependent Children in Wage-Earnings’Families; Number and Per Cent Distribution

* The children in these families are dependent nieces and nephews and therefore were not in this table classified with the other dependent children.

one dependent child, there is no information as to the relative number of families with and without such children. The tables do show, however, that families with two dependent children are the most common, as there were 112 of this type, or 24 per cent of the whole number of families. Next in frequency are the families with three children. There are 103 of these, or 22 per cent of the whole number. There are 173 families, or 37 per cent, which have more than three dependent 06145children—one family having as many as nine. When these are added to the number of those having three such children, there are 276 families, or 59 per cent of the total, having three or more dependent children.

While these families represent, in the matter of dependent children, a specially selected sample, and for that reason are not typical of the working-class population as a whole, there is no reason to think that they are

Number of Dependent ChildrenWage-Earners’ FamiliesCumulative PercentagesNone.41 or more99.62 or more83.13 or more59.14 or more37.15 or more20.26 or more11.47 or more4.1

TABLE XIX ANumber of Dependent Children in Wage-Earners’Families; Cumulative Percentages

not typical of the families having children among the unskilled laborers in Chicago. If this is the case, the fact that a large proportion of these families have three or more dependent children has great significance in relation to the adequacy of the father's earnings and of the family fund to meet the demands made upon them.

OTHER DEPENDENTS

In addition to the children there are, in 37 families, other dependents, relatives, either adults or children, who are wholly or partially supported out of the family fund. The relationship of these dependents to the chief wage-earner is very varied. He may be supporting his own mother, his wife's mother, one or more 06246nieces or nephews, a married daughter and her child, a brother or brother-in-law, when ill or out of work. In 24 cases there is only one dependent relative being supported; in 5 cases there are two; in 7 cases, three; and in in 1 case, four such dependent relatives. In 10 of the 37 cases the support is only partial as, for instance, in one Negro family where a married daughter and her child spent three months of the year 1924 with her parents. In another family, also Negro, the wife's sister and brother-in-law and their child were given a room for several months while he was unemployed. In a Polish family with two working children and one child in school, a room is given to a married daughter, her husband and child. One Italian family has with them for half of every year the man's mother, and supports her entirely during that period. The other families giving partial support to relatives are similar to these and need not be cited. It is interesting to notice the composition of the families who were supporting other dependents. This is shown in Table xx.

It might be thought that the families with several dependent children would not feel able to assume the added burden of the support of dependent relatives, but Table xx seems to indicate that this is not the case in the group which reported dependent relatives living with the family, as 20 of the 37 families reporting other dependents have three or more dependent children. It is true that 15 of the 20 are supporting only one dependent relative, but 3 are supporting two, and 1—a colored family with six children under eleven years of age—had been providing a room, rent free, for the woman's brother, his wife and child for over a year when he was unemployed. As the family is living in a five-room flat, this addition to their number, which made a household of eleven persons, is especially undesirable. Among the 06347families with fewer dependent children, 8 out of 17 are supporting more than one other dependent. The wage-earner in the one family which has no dependent children, but four other dependents, is a colored man who, with his wife who also works, supports her sister and three children. The sister makes some return by doing the housework and occasionally some laundry work. With the moneys he earns she pays the insurance for herself and her children, and in this way assures for them

Number of Other DependentsNumber of Families with Specified Number of Dependent ChildrenTotalNone123456Total372*105973112415384325113374111411

TABLE XXFamily with Dependent Children and Other Dependents

* In these families there are no dependent children belonging to the family, but there are dependent nephews and nieces.

the means for a decent burial and escape from the ignominy of burial by the county in the potter's field.

It was found also that a far larger proportion of Negro families support dependent relatives than is the case with the white families. There are 21 Negro families out of the total of 37 that have such relatives in their homes, as compared with 16 white families out of a total of 380. It is true, further, that in the colored families dependent relatives are found more often proportionately in the families with three or more dependent children than in white families of this size, and there is a large proportion of colored families who are supporting more than one dependent relative.

Boarders and Roomers

One hundred families have boarders or roomers as members of the household. For the purposes of this study it has been necessary to define these terms more exactly. “Boarders” are those who, for a certain payment every week, receive both room and board from the family; while “roomers” receive only sleeping space in return for their regular payment. Of these families, 56 are white and 44 Negro. These figures represent 14.7 per cent of the total number of white families, and 50.6 per cent of the total number of colored families. This difference is not surprising as the earnings of the Negroes have been shown to be so much lower than those of the white laborers that additional sources of income would naturally be necessary. Moreover, the fact that Negroes are charged a proportionately higher rent than white ✓people often necessitates the renting of rooms in order to meet the high rent. Table XXI shows the number of

Number of Boarders and RoomersNumber of FamiliesTotal families467Families with no boarders and roomers367Families with boarders and roomers10014422131748526 or more6Number not reported2

Table XXIFamilies with Specified Number of Boarders and Roomers

boarders and roomers living with the families. The 8 families having five or six or more roomers are all Negro families, some of whom are paying as much as $65, 06549$85, and $90 a month for six-or seven -room flats. Of the 8 families having four roomers or boarders, 5 are white and 3 Negro. The other groupings are more evenly divided between the two groups of families.

Other interesting facts are brought out by a comparison of the number of boarders and roomers and the earnings of the chief wage-earner. This is shown in Table XXII.

Earnings of Chief Wage-EarnerFamilies with Specified Number of Boarders and RoomersTotal FamiliesFamilies with No BoardersTotal with Boarders12345 or MoreNot ReportedTotal467367100442117882Not reported44368143Total Reported42333192432113582$800-$8996511$900-$99911561221$1,000-$1,09940281251321$1,100-$1,19950371333412$1,200-$1,2996344199523$1,300-$1,39963567511$1,400-$1,49981651693211$1,500-$1,5994536963$1,600-$1,6992418633$1,700-$1,799201911$1,800-$1,89966$1,900-$1,9997611$2,000-$2-0992111$2,100-$2,19944$2,200 and over11

TABLE XXIINumber of Boarders and Roomers Classified by Earningsof Chief Wage-Earner

Taking the group as a whole, the table shows that the two-thirds of the families who augment their income by taking boarders and roomers are those in which the earnings of the chief wage-earner are less than $1,500 a 06650year. Among the families in which the earnings of the chief wage-earner fall below the $1,500 line, 23.5 per cent have boarders and roomers; while among those in which the earnings are $1,500 or more, only 16.5 per cent have roomers and boarders as members of their households. It is also significant that the families having three or more boarders and roomers are in the groups where the earnings are low. Only two families having three boarders or roomers had earnings over $1,500, and none of the families with four or more had earnings higher than this level.

One other comparison, that of the number of wage-earners in the family and the number of boarders and roomers, may be of interest. This is shown in Table XXIII.

Families with Specified Number of Wage-EarnersNumber of Boarders and RoomersTotal1235 or MoreTotal families4672671434089Families with no boarders3672101133266Families with boarders10057303231442712311221145113178724842115 or more8332Number not reported211

TABLE XXIIINumber of Wage-Earners in Family Classified by Numberof Boarders and Roomers

There seems to be no relationship between the number of wage-earners in the family and the practice of taking boarders and roomers. Among the 267 families with only one wage-earner, 57, or 21.3 per cent, have boarders and roomers; and among a total of 200 families 06751having two, three, four, five, or more wage-earners, the same proportion, 21.5 per cent, have boarders and roomers.

No attempt will be made here to set our the discomforts and dangers involved in the confusing situations which result when provision is made for boarders or lodgers in homes that are already overcrowded. In a later chapter1 the facts about overcrowding are set out in some detail. The “lodger problem” must of course be visualized as one phase of the larger problem of housing and sanitation.

1 See chapter viii, “Housing and Sanitation,” pp. 104 ff. For a discussion of some of the social aspects of the problem of taking lodgers, see Breckinridge, S.P., New Houses for Old. (New York, 1921), pp. 23-32.