tion. Mention has been made of the employers who gave out-of-date lists. These caused waste of time and money, but they would not have invalidated such results as were secured. Those employers, however, who gave biased lists would have endangered, if they had not entirely vitiated, the whole investigation. In one group of lists containing 167 wage-earning women given as fairly representative of the rank and file in the establishments, a careful checking of each name with accompanying industrial data showed that, while there had been no misrepresentation in the wages actually paid, yet 70 per cent of those included were either heads of departments, assistant heads, or had had an average of 8 years' experience, ranging up to 15 years.

Such experiences made it necessary to establish a rule that all lists, with industrial data, must be taken personally by the Bureau's agents from the rolls. While these cases of bad faith were not in the majority, yet taken in conjunction with the number of wrong addresses, due to carelessness of the employees or to the carelessness of the timekeepers in taking addresses, they presented obstacles of such serious nature that another method was developed which produced equally unbiased results, and in some instances served to check the results obtained by the original method. Where permission to secure data direct from the rolls was refused by the employers, or when such rolls were under suspicion as to their completeness or currency, the names and addresses were secured through canvassing companies. The canvassers were instructed to collect lists of wage-earning women from all districts in which wage-earners lived. Such names were gathered without reference to age, experience, rank, or wage, but with constant reference to the leading industries of the city so far as women employees were concerned. Sometimes these companies secured from employers lists of names which were used "as leaders" to wage-earning districts where other names were collected.

In every case where this method was adopted the process of grouping the addresses into districts, preparatory to schedule gathering, revealed a normal proportion of names in each section of the city, except such as were occupied exclusively by wealthy residents.

PROPORTION OF WOMEN ADRIFT AND AT HOME.

To determine the proportion of women living at home and adrift, the canvassing companies' lists were used exclusively, each address being followed up and checked with reference to this subject even if no other information were obtainable except that the woman was engaged in some one of the employments within the range of this investigation. The accompanying table presents the results of the investigation in this particular for all cities.

NUMBER AND PER CENT OF WOMEN WAGE-EARNERS INTERVIEWED IN SPECIFIED CITIES WHO WERE FOUND TO BE LIVING AT HOME AND NUMBER AND PER CENT WHO WERE WITHOUT HOMES AND ENTIRELY DEPENDENT UPON THEMSELVES, OR "ADRIFT."

In

It should be remembered that the investigation was planned primarily to ascertain the actual condition of the women adrift, with special reference to their living conditions both material and moral. Early in the investigation it seemed feasible to determine the living conditions without gathering other industrial data than those bearing directly upon the income of the self-supporting women. such cities as St. Paul, Minneapolis, and St. Louis, where the investigation first started, this could be done with reasonable satisfaction because in those cities the industrial conditions are not extremely complex. In all the other cities certain industrial facts were collected as a necessary context of the living conditions. Vital information has been secured from the home girls in all cities in order that the data gathered concerning women adrift might have further illuminating context. The knowledge of the average earnings of the adrift women would lose considerable value without a reasonably clear idea of the general wage level for all women in the same employments, and particularly without some comprehension of the wage level for women living at home or with relatives.

The term "average earnings" has been used in connection with the women adrift and "wage level" in connection with the home women, not to indicate a difference in kind so much as a difference in extent of investigation for the causal facts.

The earnings of the adrift women were taken for the year or for the length of time at work in the place of present employment. The individual tabulation sheets will reveal the average length, as well as the range of service covered by the average earnings. These earnings were analyzed with great detail, showing what was the flat wage, which could be counted on with certainty; what was commission,

which is largely dependent upon character of stock and upon trade seasons; and what was overtime earnings.

Losses from the regular schedule wage were divided into losses through sickness, through voluntary vacations, and through lay-offs. There may be an apparent injustice in deducting from the schedule wage the losses through voluntary vacations, but the injustice is only apparent. As all of these adrift women are dependent upon their earnings for a living, there are few instances, as the individual city reports will show, where voluntary vacation was not taken either because the person needed the rest or because of illness among relatives.

The earnings for the girls living at home were likewise taken for the year or for the time at work in the place of present employment, but the analysis of the causes of loss of time was not so detailed as in the case of the women adrift. The ground to be covered was so great as to make it impossible to collect an adequate body of information and take the time for so lengthy an investigation for specific causes in each individual case. But there were two factors which were kept constantly in mind in getting an accurate statement of the home girl's average earnings-the girl's mental and physical equipment as a worker, and the opportunity offered by the industry upon which she was dependent for a livelihood. In other words, there was gathered up in one answer for the home girls the information that was constructed from seven answers (covering character of earnings and the specific causes of lost time) from the girls adrift. The earnings of the home girl represent a rougher estimate than those of the girl adrift.

In order that the subject might be treated comprehensively it was not only necessary to secure information concerning the level of earnings of the home girl for comparison with the average earnings of the woman adrift, but it was equally necessary to gather information concerning the drain upon the earnings of the home girl compared with the necessary and current expenditures of the woman adrift.

Even were the women wage-earners visited always willing to furnish such information as they were able, it is a difficult thing to determine the cost of living for a fixed period for the simple reason that rarely are accounts kept. The safest and most available figures seemed to be those secured by taking the weekly cost of living at the time of investigation (first making sure that conditions at that time were not abnormal). The expenditures for heat and light and laundry were secured separately and then averaged into a weekly cost for 52 weeks.

If wage-earning women (notably the American wage-earning women) were not like all other women-resentful of inquiries bear

ing on personal affairs, and particularly sensitive when the margin for incidental necessaries is precariously narrow-it would be easy to get all significant data by direct questioning. But the normal sensitiveness and the ofttimes abnormal secretiveness were no insignificant factors in the problem of this investigation. These factors show more frequently in a tendency to "gloss" the facts than in more or less brusque refusals to give information.

Inasmuch as the whole range of personal economies had to be covered in the investigation, it seemed wise to inquire first as to the facts most likely to be given with the least amount of misrepresentation; that is, to get by direct questioning such things as were possible of verification-price paid for board, room, heat, light, and laundry and from these to determine what margin was left for such necessaries as were variable and least possible of verification. It is not meant that there was no conscious or unconscious misrepresentation in the direct collection of facts. In another part of the volume will be found in detail the data for each woman visited and furnishing information. The industry, occupation, age, schooling, industrial experience, average earnings, cost of food, shelter, heat, light; and laundry, grade of housing and food, contribution to needy relatives, expenditures for car fare, amusements, etc., are given in detail for the women adrift.

In all too many instances the reader will find, between the expenditures for the current and incidental necessaries and the average earnings, little or no margin for clothes. Usually this has but one meaning: The girls have given the cost of such food as they get for themselves when other demands are not more urgent. When clothes must be purchased, when emergencies arise, something must be cut from the expenditures for the current necessaries in order to meet the demand for periodic necessaries. The information concerning the amount spent for clothes, which the margin furnishes, is effectively supplemented by a number of cases where exact records of expenditures for this purpose are given and analyzed in connection with the subject of social environment of self-supporting women. It is wholly intelligible and quite pardonable that the majority of these girls, whose earnings were inadequate, should have concealed the methods of making up the deficit. It is but the impulse of the selfrespecting to "put the best foot forward." "You see, I'm dieting, said a frail slip of a department-store girl as she held out her tray upon which the cafeteria cashier, in the presence of the Bureau's agent, put a 2-cent check, covering the cost of the girl's lunch—a small dish of tapioca. She may have been dieting, but the evidences were pathetically against the need thereof, and there were some things telling other tales to a thoughtful observer. The girl's shoes and 49450°-S. Doc. 645, 61-2, vol. 5- -2

waist and skirt were plainly getting weary of well doing, and to hold her position as saleswoman they must soon be replaced. Was she finding a way? She is given, not as an illustration of the majority, but as a type of many whose earnings are inadequate.

In a number of cases ways of making ends meet were made plain. For example: Sarah J. was first called upon in a reasonably comfortable boarding house, where she paid $4 a week for her accommodations. When the agent went a second time, in order to complete the schedule, Sarah had moved into a cheaper lodging house, where she was getting her own meals. Her own explanation was that it was time to get some new clothes and she had to "save it out of her board."

"Oh, my; where would we get our clothes if we bought meat every day?" was the way in which one of the group of four housekeeping girls answered the query as to this detail of housekeeping expenses. A woman who has spent 10 years keeping a lodging house for factory and department-store girls, not as a philanthropy but as a means of livelihood, said to the agent: "The girls' stories to the contrary notwithstanding, very few of those getting their own meals have adequate breakfasts. In some cases, of course, this is due to a desire to sleep late in the morning, but in most cases it is due to the necessity of making ends meet-when the wardrobe must be replenished, or when additional contributions to dependent relatives must be made, or doctor's bills or medicines make demands upon the meager earnings."

In some instances the reader will find in the table showing individual income and cost of living at the end of this report reference to "supplemental earnings," or "partial support." This does not mean that such girls are not practically within the ranks of the self-supporting, because the partial support is temporary and usually of insignificant amount. It is a noteworthy fact, however, that these cases of "no margin" or "apparent deficit" are not frequent, and that at least a majority of the women are living within their earnings.

How does the drain upon the woman adrift compare with the drain upon the earnings of the home girl? As has been said before, the factors in the home girl's earnings and living expenses were not analyzed in detail. In the individual tabulations for each city are set down, among other things, the earnings of each home girl and the amount of money paid to the family, either as board or as contribution. It is doubted if anything in the whole report is more significant than the large percentage of the women wage-earners living at home who were turning into the family fund all their earnings. Of the women reported in New York stores 84.3 per cent, and of those in factories 88.1 per cent, contributed all their earnings; and in Chicago and St. Louis the per cents were only slightly smaller.