■*;•?■ *•-?. !KJ m^} SQ> 3r ittp://ww\A/.archive;org/details/freldWorKSOcial gibe Centung Social Science Series FIELD WORK AND SOCIAL RESEARCH BY F. STUART CHAPIN, Ph.D. PROFESSOR OF ECONOMICS AND SOCIOLOGY, SMITH COLLEGE, DIRECTOR OF THE SMITH COLLEGE TRAINING SCHOOL FOR SOCIAL WORK Author of "An Introduction to the Study of Social Evolution," "An Historical Introduction to Social Economy," etc. ' ' \ ] » ^' V \ * NEW YORK THE CENTURY CO. 1920 *JI- Copyright, 1920, by The Centuby Co. y^^^r/ TO FRANKLIN HENRY OIDDINOS Great teacher and pioneer leader in the sdmtific ttudy of society, whose rare ability to arouse and challenge his students thought has stimulated the use of scientific method in the study of social phenomena 428631 PEEFACE A SCIENTIFIC technique has been developed in the field work of the social sciences. Scattered through the reports of private investigating agen- cies and government commissions there is a vast amount of valuable information on methods of making scientific investigations of social condi- tions by first hand contact with the facts — that is, by field work. This book gathers together the well tested methods and techniques and attempts to present them in an accurate and practical form. The treatment is not exhaustive, yet it is believed that all of the significant points are touched upon with sufficient detail to make the book useful as a manual to field workers. The theoretical princi- ples underlying scientific method applied to the investigation of social conditions are outlined with sufficient thoroughness to make the book sugges- tive to students. The methodology of field work is developed in chronological order as the reader follows the suc- cessive chapters of the book. The material is, however, so arranged in chapters and under text headings and by index, that the reader will not have to wade through a mass of descriptive mate- rial to find some detail of technique in which he is especially interested. PEEFACE Actual field work investigations of many differ- ent kinds are described in considerable detail and the theoretic principles underlying procedure are so stated that the practice may be critically exam- ined in the light of well-established methods. F. Stuart Chapin Northampton, Mass. May, 1920 CONTENTS PART I THE APPROACH TO FIELD WORK CHAPTER PAGE I The Place op Field Work in Social Research 3-18 Science Is Impartial 4 The Tasks of Science 4-5 The Inductive Method 6-11 Field Work and the Inductive Method . 11- 16 Field Work in Social Research . . . 16- 18 Selected References ........ 18 II Critical Examination of Document- ary Sources Precedes Good Field Work 19-45 The Sceptical Attitude Toward Docu- ments 19-20 The Historical Method 20-21 External Criticism 21- 24 Internal Criticism 24r- 26 The Author's Good Faith and Accuracy 26- 36 Summary of Principles of Criticism . 36- 38 Partial List of Common Documentary Sources 38- 45 Selected References 45 PART II SCOPE AND ORGANIZATION OF FIELD WORK III Types op Field Work and Their Prob- lems 46- 72 The Three Types of Field Work . . 46- 47 Problems of Field Work 47-51 X CONTENTS CHAPTER PAGE The Investigation of Industrial Relations 51- 54 Health Investigation 54- 65 Infant Mortality 65-71 Selected References 71- 72 IV The Investigative Procedure of Case Work 73-97 The Underlying Logic of Procedure . 74- 79 The First Interview 79- 81 Sources Outside the Family .... 81- 87 Record Writing 87-89 Examples of Medical-Social Case Inves- tigation 89- 97 Selected References 97 y Samples that Are Representative — The Partial Canvass — Investigative PROCEDimE OF the Socml. Survey . 98-126 The Social Survey 99-100 Survey of Syracuse, New York . . . 100-105 Survey of Springfield, Illinois . . . 105-107 Methods of Social Surveys .... 10Z=vll2 Examples of Representative Sampling . 112-116 The Theory of Inductive Inference . . 116-118 The Theory and Practice of Random Sampling 118-121 Empirical Rules for Representative Sampling 121-126 Selected References . . . . . . 126 VI Complete Enumeration of a Govern- ment Census — The Technique of Full Cajntvass 127-147 Purpose and Scope of the Census . . 127-129 Massachusetts State Census of 1915 . 130-132 Preparation for Field Work .... 132-139 Narrative 139-142 Supervision of Field Work .... 143-147 Selected References . . . [.^ . . 147 CONTENTS PART III /^ SPECIAL PROBLEMS CONNECTED WITH FIELD WORK CHAPTER PAGE VII Purpose and Preparation of Schedules FOR Field Work 148-192 The Schedule a Mechanical Aid to Ob- servation 149-156 The Form of the Schedule .... 156-164 Arrangement of Inquiries .... 164-167 Content and Phrasing of Inquiries: Definition of Units 167-172 Content and Phrasing of Inquiries: Phrasing of Questions 172-175 Quantitative Expression in the Schedule 176-185 Summary of Principles 186 The Schedule as Determined by Practice in Tabulation 186-187 The Questionnaire Method .... 187-191 Selected References 191-192 VIII Editing, Classification, Transcribing, Tabulation and Interpretation of Field Work Data 193-219 Editing Schedules 194-197 Classification 197-203 Transcribing 203-205 Tabulation and Interpretation of Data . 205-218 Selected References on Statistics . . 218-219 Index . . . , 221-225 LIST OF ILLUSTRATIONS 151, 1. Lot card used in housing investigation , 2. Lot card used in housing investigation . 3. Family schedule used in investigation by Health Insurance Commission of Illinois 4. Schedule used in investigation of infant mortality by U. S. Children's Bureau . 5. Population schedule — male card . 6. Population schedule — family card . 7. Housing schedule — ^house card . 150- 8. Multiple house card .... 9. Portion of a questionnaire . 10. Commons' dwelling house score card 11. Score card for manner of living . 12. Questionnaire sent to employer . 13. Questionnaire sent to labor unions 14. Census punch card 15. Tabulation. Arrangement A . 16. Tabulation. Arrangement B . 17. Frequency table, incorrect arrangement 18. Frequency table, correct form . PAGES 12 13 56- 58 68- 69 134 135 152-153 158-160 165 178-181 184 189 190 206 209 210 216 216 PAET I THE APPKOACH TO FIELD WOEK FIELD WORK AND SOCIAL RESEARCH CHAPTEE I THE PLACE OF FIELD WORK IN SOCIAL EESEAECH Disinterested examination of contemporary so- cial facts is rare. Only in recent years has the study of social conditions begun to be scientific instead of sentimental. The very complexity of causation which lies back of social problems has often discouraged painstaking analysis and defied systematic investigation. Prejudice and super- stition still hamper and frequently prevent an im- partial examination of things as they are. Al- though, as Keller^ says, *^A man can count the legs of a fly and report his findings without hav- ing his heart wrung because there are too many or too few,'* when it comes to social facts, disin- terested study is difficult because of the strong emotional coloring of everything that is estab- lished and traditional. Controversies about so- cial problems usually give rise to more heat than light. 1 Keller, A. G.— "Sociology and Science," The Nation, vol. 102, No, 2653, p. 475. 3 '^: ^''PIELD WGiiK AND SOCIAL KESEARCH SCIENCE IS IMPAETIAL Scientific study of social conditions is needed to discover truth. The scientific student of contem- porary social relations should strive to brush aside these obstacles of emotional bias and objectify the approach to his problem by eliminating self from his judgments. He should provide an argument which is as true for every other mind as for his own.^ It should be possible for any observer or experimenter to arrive at the same results, pro- vided only that he follows the same method. The classification of data should also be independent of the individual thinker. It is in these things that the element- of universality which character- izes modern science resides. It is this all-embrac- ing, this unrestricted adaptability of science which makes it impersonal and almost devoid of individual whim, or bias or prejudice. THE TASKS OF SCIENCE Science has three tasks: first, the discovery of the laws of natural phenomena — accomplished by the use of the inductive method; second, the dis- covery of causes — accomplished by hypothetic in- ference; and third, the prediction of effects — ac- complished by the use of deduction.^ Science is judged by its success in the prediction 2 Pearson, K. — Grammar of Science, 2nd Ed. 1892, p. 6. sPeirce, C. S. — "A Theory of Probable Inference," Studies in Logic, Johns Hopkins University, 1S83. THE PLACE OF FIELD WORK 5 of effects. In the realm of physical phenomena the laws of matter and motion, the laws of chem- ical attraction and of combination, have been made the basis of remarkable practical achievements in the applied sciences of engineering and sanita- tion. On the other hand in the social field its students have signally failed to predict effects. The reason for this seems to be that only recently has the inductive method been used in the study of social phenomena. Semi-philosophical gener- alizations have been hastily formulated without contact with the facts and from these glittering generalities deductions have been made with scant results. Impatiently turning aside from pains- taking accumulation of facts, social *^ science'' has inclined to the ** painful elaborations of the obvious. ' ' Says Pearson,* ^^The unity of science consists in its method, not in its materials,'' and again, **It is not the facts themselves which make science, but the method by which they are dealt with." The striking surface contracts in material — plan- ets, bacteria, beetles, men, mice, elephants, and all the rest — catch our attention and turn it from the central feature of inductive method which is everywhere the same. Whoever uses this induc- tive method is scientific no matter what material it is that he studies, nor how obscure the comer in which he works. * Pearson, op. cit. 6 FIELD WORK AND SOCIAL RESEARCH THE INDUCTIVE METHOD The uniform steps of the inductive method of modern science are as follows : (1) The working hypothesis. The scientist does not go out and make random and haphazard observations of all phenomena. He limits his field by adopting provisionally some hypothesis which will provide a systematic basis for selecting his material. This brings his problem within manageable proportions and saves time. Now hypotheses are often suggested by analogy, and analogies are proverbially dangerous; but the hazards inherent in the analogical method disap- pear when every hypothesis is subjected to the acid test of facts. (2) Collection and recording of facts of ob- servation is the second step in inductive method. The approach of the student to his facts should be objective. The effort to be disinterested should eliminate personal bias. There should be method, system, orderliness, in observing social phe- nomena. Standardization of methods of observa- tion and recording by permitting the comparison of observations of different students made at dif- ferent times and in different places, contributes to scientific progress. (3) Classification of the recorded facts of ob- servation into series and sequences throws light on the natural relationships among these facts. This is the third step of the inductive method. THE PLACE OF FIELD WORK 7 (4) Discovery of some short formula or law to explain the sequence of facts and to express their relationships is the concluding step of the induc- tive method of science. Deductions from induc- tive generalizations established in this way form the reliable predictions of science. In the oldest of sciences — astronomy — the in- ductive method has been used with remarkable success. The position of a certain planet was carefully observed on a certain day and the result recorded. The observation was repeated at in- tervals on various subsequent days. When these observations were classified a record of the suc- cessive positions occupied by the planet during its course was finally obtained. It remained to dis- cover the formula which explained this phe- nomena. Kepler discovered that if it is assumed that each planet moves around the sun in an el- lipse an explanation is provided which is in pre- cise agreement with the observed facts. Suppose we observe the planet's position on January 1st. Knowing its rate of motion we could calculate the position it would occupy by January 14th by de- termining how far along the ellipse it would have moved by that date. When January 14th arrived we should find that the calculated position and the position determined by actual observation agreed precisely. Since this test is invariably satisfied, Kepler 's formula has been accepted as a scientific law. In other words, it satisfies the test of sup- plying reliable predictions of future events. 8 FIELD WORK AND SOCIAL RESEARCH Although great precision still lacks in the field of life sciences, certain laws have been discovered which enable us to predict effects within a certain margin of error. The monk, Gregor Mendel, has given us his law of heredity as a consequence of painstaking experimentation in plant hybridiza- tion carried on with the conunon garden pea. He first determined what characters were constant for certain varieties and species and then pro- ceeded to cross one variety with another. Cross- ing was accomplished by dusting upon the stigma of one variety, the pollen of a different variety. In every case he discovered that the plant that developed from such a cross exhibited only one of the two contrasting characters of the parent plants. By crossing yellow-seeded and green- seeded plants he obtained in the next generation by self-fertilization, 6,022 yellow seeds and 2,001 green seeds, or about three yellow to one green. Crosses of round- and wrinkled-seeded varieties yielded in the hybrid generation, 5,474 round and 1,850 wrinkled seeds, or again, the ratio of three to one. The hybrids of tall and short parent plants, produced on self-fertilization, 787 tall stemmed plants and 277 short stemmed plants. Mendel found the same proportions held prac- tically constant for other characters. One trait dominated the other in the hybrid generation, and then persisted in the second filial generation in the ratio of three to one. Thus by patient observa- tion and classification of the facts, Mendel discov- THE PLACE OF FIELD WORK 9 ered a principle of inheritance. His explanation, or theory, held that the reason for the splitting of pure dominants and pure recessives from hybrid parents must be found in the composition of the male and female sex cells. If it is assumed that the germ cells are pure with reference to the constant character observed in the original gen- eration, then the hybrids possess germ cells half of which are pure for one character and half for the other, and it follows that self-fertilization in this generation w^ill produce, on the average, three plants showing the dominant character to one with the recessive. Since this explanation agrees with the facts and, within the margin set by the law of averages, enables us to predict the conse- quences of crossings of known contrasting traits, a scientific law of heredity of living things has been formulated. There is considerable evidence to show that in man^ such characters as short fingers and toes (brachydactyly), webbed fingers and toes (syndactyly), and supernumerary digits (Polydactyly) are dominant over the normal con- dition, and that in the nervous system, hereditary feeble-mindedness, hereditary hysteria, heredi- tary epilepsy and so on, are recessive to the nor- mal. In so far as studies of this sort are sub- stantiated by additional evidence, the enormous importance of MendePs law for the control of human welfare becomes evident. 5 Conklin, E. G.— "The Phenomena of Inheritance," The Popu- lar Science Monthly, vol. 84, no. 22, pp. 440-441. 10 FIELD WORK AND SOCIAL RESEARCH The German statistician Ernst Engel, gath- ered together data showing the expenditure of Saxon working-class families. He classified these figures as expenditure for food, rent, fuel and light, clothing and sundries. Study of this data led him to formulate the following inductive gen- eralizations : first^ the greater the income, the smaller the percentage outlay for subsistence; second, the percentage outlay for clothing is ap- proximately the same, whatever the income ; third, the percentage for lodging or rent and for fuel and lighting, is invariably the same, whatever the income; and fourth, as income increases in amount, the percentage of outlay for sundries be- comes greater. Students of family budgets and the standard of living in America have confirmed the first and fourth of his conclusions; but the second does not hold good, since expenditures for clothing usually rise with increase in income; while the third is only partially true, for the per- centage of rent varies only slightly as income rises, that for light even less, and expenditure for fuel actually falls. Conditions making for greater elasticity and freedom in expenditure of American families probably explain this disagree- ment. Recently Ogbum has given mathematical- statistical expression to these principles of income and expenditure thus stating them in precise quan- titative terms.^ Although EngePs ^4aws of in- 6 Ogburn, W. F. — in Quarterly Pub. Amer. Statistical Assoc, vol. 16, no. 126, June 1919, p. 374. THE PLACE OF FIELD WORK 11 come" are not yet admitted to the rank of scien- tific laws, considerable progress in that direction has been made and the validity of the inductive method has been established in one of the most perplexing corners of the social-economic field. Observation of natural phenomena under con- ditions of control has been the secret of the suc- cess of physical science. The experimental method is a tool of extraordinary efficiency. But the experimental method is extremely difficult of application to social phenomena.'^ The number and interplay of factors in any social problem make it almost impossible to determine all of the agents that are at work, and until the factors can be defined it is not possible to control all condi- tions save the one to be measured — and yet this is the sine qua non of the experimental method. Progress in social science is therefore hampered by the difficulty of making observations under conditions of control, and this fact explains in considerable measure its slow development as compared with the brilliant achievements of phys- ical science. FIELD WORK AND THE INDUCTIVE METHOD Although it is difficult to isolate one factor at a time in the study of social phenomena, there is no excuse for failure to make direct observations of social facts. Now it is just at this point in social f Chapin, F. S. — "The Experimental Method and Sociology," Pop. 8ci. Mo., vol. 4, nos. 2 and 3, 1917. IS 14 FIELD WOKK AND SOCIAL RESEARCH research that field work supplies the technique for observing social phenomena by first hand contact in the field. Field work is the technique of study- ing social conditions by first hand or direct con- tact with the facts. Systematic field work sup- plies the means for approaching the study of social phenomena in an objective way. By the use of carefully prepared schedules the personal equation of the field worker is minimized and comparable observations are obtained. The chief mechanical tool of field work is thus the schedule. In so far as the plan and arrangement of inquiries on the schedule objectify the study of social con- ditions by supplying the basis for quantitative ex- pression of qualitative facts, the schedule is an instrument of scientific observation. In this sense the schedule is the analogue in social sci- ence of the telescope, the camera or the spectro- scope in the science of astronomy. These latter instruments of observation extend the power of the human senses and permit the recording of observations in an objective fashion. As we shall see in chapter vii, the schedule used in field work performs the same function in a limited way. Systematic field work, methodical observation of social facts, requires careful organization of the investigation. In the following chapters we shall describe and illustrate the different methods of planning the field work of investigation, involving the principles guiding the selection "and instruc- tion of the field staff, the preparation and use of THE PLACE OF FIELD WORK 15 schedules, the supervision of workers in the field, and finally, the editing of schedules turned in by field workers. Field work however well-planned is not the first step in the scientific investigation of social conditions. It must be remembered that field work is the technique of making direct observa- tions by first hand contact with the facts. Sup- pose some one else has gathered the data required by the investigation? Before plunging into field work therefore, it is necessary to survey the sub- ject of investigation to discover whether some one else had not already gathered the necessary data. This means that documentary records or written descriptions and reports of former studies should be consulted before planning or starting the field work phase of the investigation. It is commonly supposed that the scientific chemist spends most of his time experimenting in the laboratory. This is not the case, for three-quarters of his time in any research work is devoted to reading at the library all the literature on the particular subject he is investigating. In reality, only one-quarter of his time is put in at the laboratory in actual chemical experimentation. Similarly the social scientist before he begins field work should thor- oughly survey the field to discover what has al- ready been done. Thus indirect observation pre- cedes direct observation. Since the systematic use of documentary sources is a subject in itself and one in which a real scientific technique has 16 FIELD WORK AND SOCIAL RESEARCH been built up by historians, we shall treat of it iu some detail in chapter ii. A clear undestanding of the place of field work in scientific method applied to the study of social conditions is necessary if we are to orient our- selves properly with reference to this specialized branch of social investigation. We have, there- fore, considered field work in the light of the three steps of the inductive method and found that field work, being a technique of directly observ- ing social facts, corresponds to the second step of the inductive method, e.g., that of collecting and recording the observations of natural phenomena. But we have discovered that a survey of docu- mentary records of observations previously made comes first, and is only followed by field work when it is certain that existing records are incom- plete. It is thus clear that field work and the historical method (critical use of documentary sources) are the two specialized techniques of social science for collecting and recording the ob- servations of social phenomena. FIELD WOKK IN SOCIAL EESEAECH While we are considering this matter it is in- teresting to inquire whether in social science there has been developed any method or technique cor- responding to the third and leading up to the fourth step of the inductive method? The statis- tical method does in fact furnish a highly efficient tool for scientific classification (tabulation) of THE PLACE OF FIELD WORK 17 social data and for the interpretation (graphic methods, ratios, averages, index numbers, cor- relation, etc.) of social and mass phenomena. It is the statistical method, therefore, which sup- plies social science with a special technique for classifying and interpreting social data. The place of field work in the procedure of social re- search is thus intermediate between the applica- tion of the historical method and the statistical method. If we should attempt to enumerate the methods of social research, we should find that there are three distinct methods, each with a highly developed technique all its own: (1) the historical method (developed by historical stu- dents) ; (2) field work (developed by statisticians and social workers) ; and (3) the statistical method (developed by statisticians and mathe- maticians). The inductive method Methods of sooial research 1. The working hypothesis. 2. Collection and recording of 1. The historical method of facts of observation* critically using document- ary sources, (indirect ob- servation ) 2. Field work. Observation by first hand contact with the facts, (direct observa- tion ) a. Case work b. Sampling c. Complete enumeration. 3. Classification of the facts of 3. The statistical method observation into series and sequences for comparison. 4. Generalization from these classified facts to some short formula or law which explains their rela- tions. a. Tabulation b. Graphs, ratios, aver- ages, indexes, correla- tion coeflScients, etc. 18 FIELD WORK AND SOCIAL RESEARCH The above tabular arrangement may serve to clarify the relations of the different methods of social research as they form special phases of the inductive method in social science. SELECTED REFERENCES Bowley, A. L. — The Measurement of Social Phenom- ena, 1915, Ch. 1. Chapin, F. S. — ''The Experimental Method and Sociology/' The Popular Science Monthly, vol. 4, nos. 2 and 3, Feb.-March, 1917. Duncan, C. S. — Commercial Research, 1919. Giddings, F. H. — Inchictive Sociology, 1902. Jevons, W. S. — Principles of Science, ISl 4:. Mees, C. E. K. — The Organization of Industrial Scien- tific Research, 1920. Pearl, R. — Modes of Research in Genetics, 1915. Pearson, K. — The Grammar of Science, 1892. Peirce, C. S.— ''A Theory of Probable Inference," Studies in Logic, Johns Hopkins University, 1883. Westaway, F. 'W .—Scientific Method, 1912. CHAPTER n CRITICAL EXAMINATION OF DOCUMENTARY SOURCES PRECEDES GOOD FIELD WORK The first step in making any investigation is to discover what others have done in the same field. This means careful examination of docu- mentary sources and written records to be found in libraries. It is short-sighted and wasteful of time to plunge at once into field work. THE SCEPTICAL ATTITUDE TOWARD DOCUMENTS In the use of all documentary sources there is a fundamental need of scepticism. The investi- gator has a spontaneous tendency to yield assent to affirmations made in documents and to repro- duce them without distinguishing them from the results of his own observations.^ Since credulity is natural and deeply rooted in human indolence there is need of criticism of documentary sources. It is far easier to admit a statement than to criti- cize it; it is much simpler to believe than to dis- cuss; and less difficult to accumulate evidence, documentary statements, facts and quotations, than to weigh them. 8 Langlois, C. V., and Seignobos, C. — Introduction to the Stiidy of History (G. G. Berry translation) 1912, p. 69. 19 20 FIELD WORK AND SOCIAL RESEARCH Every student of social conditions should try to overcome his intellectual inertia. He should train himself not to accept indiscriminately, and without any attempt to verify, hearsay reports, anonymous statements, and documents of doubt- ful authority. By dint of continued and consci- entious practice he should form the habit of criti- cism. It must always be remembered that documents supply indirect, as contrasted with direct knowl- edge. Documents are the records of the thoughts and actions of other men in a recent or a remote past. They are merely traces of psychological operations once performed. Now events are known to us in two ways only: (1) by direct ob- servation while they are in progress (this is the sphere of field work) ; (2) or indirectly, by the study of traces which they have left behind them. The document is thus the record of an observa- tion. In using documentary reports of events or descriptions of conditions, we do not observe our- selves but draw inferences from the observations of others. The documentary source is therefore, merely the starting point, with the fact as the goal. Field work, on the contrary, begins with first hand observations of the facts. THE HISTOEICAL METHOD Historians have developed a scientific technique for the critical use of documentary sources. Al- though some phases of the historical method are EXAMINATION OF SOURCES 21 too specialized to be of service in the ordinary social investigation, the larger part of the histori- cal method is of direct applicability to the prob- lems with which the student of social research deals. Indeed, the historical method is one of the three important methods of social research, the others being, field work and the statistical method. Now the main divisions of the historical method are two:^ (1) external criticism of the objective characteristics of the document, including critical examination of authorship, critical classification of sources, and criticism of the form or text of the document; (2) internal criticism of the sub- jective characteristics of the document, involving a consideration of the real meaning of the author 's statements, his good faith and accuracy. EXTERNAL CEITICISM The form and appearance of the document as distinguished from its contents should first re- ceive our critical attention. Here again, we must be on our guard against the spontaneous ten- dency of the human mind to place confidence in the indications of authorship. The ^ impulse of confiding trust *^ must be checked. We should obtain satisfactory answers to such questions as : where and when did the document originate? Who was the author? What were the sources used? To be really critical, a classification of sources e Ibid., pp. 66-69. 22 FIELD WORK AND SOCIAL RESEARCH should show discriminate selection of material and absence of credulity. Compiling a critical bibliography is one of the first things to do in starting an investigation. There are different methods of classifying documentary sources. Certain methods should not be followed, for ex- ample: do not make notes on documents read or consulted and enter them in a notebook one after another in the order studied, because such a pro- cedure defeats classification ; do not enter notes in a book under headings which form a prearranged scheme, for s&ch a procedure leads to a system of classification that is too rigid; it is still worse to rely on memory and fail to make notes at all. The best plan, on the whole, is to make loose leaf notes as the examination of documents proceeds. All slips or cards should be of uniform size with a heading upon each to identify the subject of its contents, and a full citation to source at the bot- tom. Slips or cards prepared in this way may be filed, classified and cross-indexed for ready refer- ence. Eeferences to sources should follow con- sistently one of the standardized methods of foot- note citation. Author, title, date of publication, number of edition, volume, part, chapter and page, should be given to properly indentify the precise source. It is only in this way that your findings are convincing to other minds, for others may then check up your evidence. Historians have developed certain principles to test the relative value of documentary evidence EXAMINATION OF SOURCES 23 when, (1) the original document is preserved, (2) when a single copy only is preserved, and (3) when several copies are preserved and errors must be compared. These principles are too spe- cialized to discuss here, but in general it may be said that the document of original entry is more reliable than the copy or derivative. It should also be remembered that errors in the original document survive in the copy. Criticism of the external characteristics of documents, outlined in the preceding paragraphs, has its limitations and its dangers.^^ The vota- ries of critical scholarship are as much in error in unduly exalting the merits of criticism as are su- perficial and sentimental persons who hold criti- cal scholarship in contempt. It should be remem- bered that criticism of the externals of a docu- ment is merely preparatory to the more difficult and important task of criticizing the psychological aspects of the source. External criticism is thus a means to an end, and not an end in itself. While critical analysis into the minutiae of docu- ments satisfies the impulses towards collecting and puzzle-solving, it sacrifices the higher facul- ties to purely critical learning. There is danger that the pursuit of this branch of criticism will lead to dilettantism and to hypercriticism. The technique of criticism tends to become more im- portant than the results. The tool becomes the goal. Criticism comes to exist for the sake of 10 Ibid., pp. 114-134. 24 FIELD WORK AND SOCIAL RESEARCH criticism. The result is loss of power to work. *^Some of the most accomplished critics merely make a trade of their skill, and have never re- flected on the ends to which their art is the means." ^^ INTERNAL CRITICISM Internal criticism of the subjective characteris- tics of a document is the more important part of criticism. In order to determine what in the doc- ument may be true it is necessary, strictly speak- ing, to trace in each case the mental operations which began with the observation of the fact and ended with writing the words in the report, fiow- ley ^2 says that it is a good plan before evenYead- ing a statistical account to *^sit down and think quietly what statistics ought to have been col- lected, if possible, for the purpose in hand, and what sources of information exist, or should ex- ist.'' In the case of a wage study the weekly rate, supplementary earnings of other members of the man's family, allowance in annual earnings for periods of unemployment, and so on, should be considered. Since in most cases an analysis of all the men- tal operations of the author is out of the question, psychological criticism concentrates on two lines of examination: ^^ (1) analysis of the contents of the document to ascertain what the author meant ; 11 Ibid., p. 143. i2Bowley, A. L. — Elementa/ry Manual of Statistics, 1910, p. 67. isLanglois, op. cit., p. 143. EXAMINATION OF SOURCES 25 and (2) analysis of the conditions under which the document was produced in order that the author's statements may be verified and evaluated. Un- less a document is critically studied in this way there is danger that the investigator using it as a reference, will read into the text his own opinions — especially in cases where the author's language and thought differ from his own. A system of loose leaf notes or slips or cards is the most helpful mechanical aid to this analysis. Each slip should indicate for the part of the docu- ment cited, the general sense of the text, and the object and views of the author. It should always be remembered that the literal meaning of the author's language is not a fixed quantity. Such facts as the time in which the document was written, the language of the coun- try, the author's own manner of using language and the general sense of the context all constitute variable factors. In using documents written in a foreign language these considerations apply with special force. In general, however, it is neces- sary to examine in this critical way only those ex- pressions which, from their nature, are liable to take on different meanings, such for example as, classes of men, institutions, feelings, customs, common objects, terms and units used in social- economic investigations. Does the term ** births" used by the author include or exclude still births? Does ** foreign population" mean foreign born, native born of foreign parents, or both ? 26 FIELD WORK AND SOCIAL RESEARCH The real meaning of the author 's language may be disguised by jests and hoaxes, in allegory, sym- bolism, allusion, implication, and in such ordinary figures of speech as metaphor, hyperbole and litotes. To detect the existence of an oblique sense in the author's expressions, there should be evidence of absurdity, incoherence, contradiction or obscurity in the literal sense of his statements. Having determined what the author meant we must proceed to a critical estimate of his good faith and accuracy. Owing to the fact that state- ments in documents are found to contradict one another, it is necessary to examine written rec- ords carefully to eliminate errors. The high de- gree of vitality possessed by spontaneous credu- lity is shown in the common tendency to accept as true every written statement, as if no author ever lied or was deceived. The scientific student should cultivate the habit of methodical doubt of documentary statements. Although a document may be quite authentic as to origin, this does not in the least establish any presumption in favor of the truth of the statements it contains. As writ- ten sources are used, care should be taken by the reader to analyze and criticize each statement. This means analytical criticism of the document and not criticism en bloc. "While our description of this procedure sounds as if it were too slow and intricate to be practicable, this is really not EXAMINATION OF SOURCES 27 the case. Methodical doubt consists in the hahit of performing certain acts of thought. As soon as the initial difficulty is overcome the habit is readily established and methodical distrust and criticism become second nature — performed with- out consciousness of disagreeable slowness or dif- ficulty. Beware of putting faith in the form in which a statement is cast. Form does not indicate sin- cerity or accuracy of the author. The so-called *^ accent of sincerity" presented by a statement is an illusion. Vehemence in affirmation does not necessarily mean strength of conviction. It often indicates the reverse. Profusion and precision of detail are in themselves not a guaranty of ac- curacy in facts, however vivid may be the impres- sion they produce upon the reader. The value of the author's statement is deter- mined by the conditions under which he made his observations. Critical investigation of author- ship begins, therefore, with the preparation of a general set of questions which have reference to the possible causes of falsehood. The document is then tested against these questions to discover those causes which may have rendered the au- thor's mental operations incorrect and hence viti- ated the results. Criticism of the particular statements contained in the document is carried out by testing each statement against a second list of questions which relate to the causes of inac- curacy characteristic of mental processes. 28 FIELD WORK AND SOCIAL RESEARCH Are there reasons for doubting the author's good faith because he was placed in one of those situations which ordinarily incline a man to be in- sincere? It becomes, therefore, a question of motives, and we ask, had the author an interest at stake, did he seek to gain practical advantage for himself by the statements he wrote down! If so, he had an interest in deceiving. This is the case with most official documents. We must ask what interest the author could have thought he had in misrepresentation. We find the answer in his ideals and tastes. Now it should be remembered that instead of individual interest, the author may have sought to serve some collec- tive interest of a political party, an economic class, or a religious denomination. Much that is given in the United States Senate's report on *' Prices, Wages, and Transportation'' of 1893, appears to be susceptible of this interpretation.^'* The motive to show the beneficient effect of a Ee- publican tariff on prices, wages and transporta- tion, seems to supply an explanation of the un- representative character of the data upon which the final indexes in this report were based. Of 21 industries investigated, 11 were represented by one establishment, each. Although there were 353,444 clerks in the dry goods business, the index for wages in the dry goods business was based on the wages of employees in one store in New Hamp- 14, 15 and 16, see respectively vol. I, p. 175; vol. Ill, pp. 857-863; and vol. II, p. 313. EXAMINATION OF SOURCES 29 shire, where one porter, eight salesmen, and six salesladies, were employed.^^ No investigation was made into the wages of agricultural laborers, although the United States is largely an agricul- tural country. Of course their wages would have been low and had they been incorporated into the material upon which the index was based, the lat- ter would have been less high. The wage index for brewers ^^ in the industrial class, *^Ale, beer, and porter, ' ' was based on one brewer in one New York establishment, and weights were not used. The question of good faith deserves special con- sideration wherever newspapers are used as sources. The partisan character of editorial col- umns is well kno^vn. Editorial policy is con- trolled by some individual or perhaps by some corporation. Advertisements may supply a clue to the nature of this influence. The character and frequency of certain types of advertisements appearing in a given newspaper indicate the kind of commercial patronage it receives and may sug- gest the source or explain the significance of edi- torial bias. In the news columns, headings are often chosen, that consist of words or phrases ap- pearing below but which, when removed from their context misrepresent the real meaning of the re- port or distort the evidence in the news columns. In other cases wilful misrepresentation is resorted to. The heading, *^ Wilson opposes military training'^ (in large type), with the modifying phrase, ^* President, in letter to Secretary Baker, 30 FIELD WORK AND SOCIAL RESEARCH gives reasons for objecting to house democrats making it an issue'' (in small type), gives the reader the impression that the President's letter was written to oppose military training. The truth in the case, is however, quite the contrary, for in the middle of his letter the President makes this statement, *^ . . the moderate and carefully conducted course of military training may have the highest possible advantages," and then pro- ceeds to say that the principle of moderate mili- tary training to which he had given his approval should not be made a political issue. The news- paper in which these headlines appeared was at- tached to the most narrow interpretation of Re- publican party politics. Was the author placed in a situation in which ■he was forced to tell an untruth? This situation .^-^exists wherever a document has to be drawn up in conformity to rule or custom. In modified degree ^--' this question applies to many legislative records of national, state or municipal origin. It should be remembered that statutes or ordinances are records of a compromise of differing opinions on an issue. They should be examined in the light of written records of the discussion which pre- ceded them. This is to study the process by which the compromise was reached, and throws light on motives underlying the action taken. Minutes of the proceedings may be examined and it may sometimes be wise to go back of these to records of special committees in order that their rules of I EXAMINATION OF SOURCES 31 order may be understood. All this testimony should be checked against the general rules of evi- dence. Other questions may now be asked. Was the author influenced by sympathy or antipathy and hence biased to such an extent as to distort facts in representing his opponents in an unfavorable and his friends in a favorable light? Was the author influenced by vanity to violate the truth? Perhaps certain statements in the text were made with a view to impress the reader with the im- portance or power of the author or the group he represents. Are the statements of the author in- fluenced by desire to please the public? Is there over-deference to public opinion which leads to distortion of facts? Answers to these questions depend upon the particular public grotip to which the document is addressed and by the special morals or manners of this public. Is there sus- picion of dramatic or literary distortion in the statements contained in the document? Having examined the author's good faith by obtaining an answer to these questions, we must now consider the reasons for doubting the accu- racy of the author's statements. In some cases the author was a poor observer because of forces of which he was not aware such as hallucinations, illusions, mental defects, or prejudices. Do these considerations apply in the document under examination? The best that we can do in answer to such a question is to learn 32 FIELD WORK AND SOCIAL RESEARCH from information derived from other sources, or by comparison, whether the author had a general tendency to this sort of error./ Experimental psychology has demonstrated by laboratory tests ' ' that an errorless report is not the rule and that attention does not guarantee accuracy.^^j Of course, the inquiry into prejudices overlajTs our previous questioning of motives for falsehood. Yet in this way evidence as to good faith and ac- curacy becomes cumulative and corroboratory. Finally, it should always be ascertained whether the author has put forth the statement in answer to a question. If this is the case, then we should carefully examine the situation to determine the extent to which the statement as an answer to a question was distorted by a desire to please the interrogator, and by the natural tendency of ques- tions to suggest their own answers. Was the author badly situated in time and place to observe? The ideal conditions of observation are those in which the observer, without any pre- conceived idea about the result, was placed where he could see correctly and recorded the observa- tion immediately in a precise system of notation, accompanied by precise indication of the methods used.^^ Since chances of inaccuracy are always present it is necessary to secure an answer to the foregoing question. Perhaps the author was 17 Whipple, G. M. — Manual of Mental and Physical Tests, 1910, pp. 286 et seq. 18 Langlois and Seignobos, op. cit., p. 174. EXAMINATION OF SOURCES 33 present at the event, or did really observe the con- ditions on the spot. Yet even so, failure to re- cord the observation immediately and in precise language may invalidate conclusions based upon his report. We should be careful to distinguish between the author who is a mere witness of the event which he describes and one who is a trained observer. The author who is a scientific observer proceeds by fij^ed rules and records his observa-' tions in language of rigorous precision; whereas the author who is a mere witness, observes with- out method, and reports in unprecise language, moreover, we do not know what precautions he has taken to make his observations accurate. Most newspaper reports are of witnesses in this sense of the word and not reports of observers. They are not reliable because first impressions and hearsay play such a part in them. The garbled reports of public speeches and addresses are a case in point. There are other questions to be answered. Was the author at all negligent or indifferent in making his observation? Perhaps from idleness or negligence his report of the event distorts the facts. It is common for reporters to publish ac- counts of gatherings they never attended. Was the fact reported of such a nature that it could not be directly observed? This is the case with statistical totals or comprehensive judgments which are propositions derived from observations er of family under both life and sickness insurance. Include insurance at present carried and in- surance carried at time of death for members of the family who have died in the past 12 months. Care should be taken to get accurate information on these questions. Amount. If children's insurance varies with age and length of time policy is held, give minimum and maximum. Isame of Company. If there is not space to write out name, abbreviate and explain abbreviation in footnote except for well- known companies, e.g. "Prudential," "Metropolitan." Under sickness insurance, weekly benefit means the benefit called for by the policy not the benefit actually received. Enter the amount of cash benefit and add "And medical attention" or what- ever is necessary to indicate the rights of the insured. If more space is needed write "note" and describe fully at end of schedule or bottom of page. 20. Include still births, miscarriages and abortions. Enter "P" or "M." If nursing care given by association, e. g. "\rNA" — specify. Enter name of hospital. 61 €2 FIELD WORK AND SOCIAL RESEARCH 372 cases had to be thrown out because of incom- plete information, moving away or family disin- tegration, leaving 628 cases finally available for intensive study. The same schedule was used in all three divi- sions of the investigation. This schedule was de- veloped after an examination of schedules used in family studies made elsewhere. A trial schedule for experimentation was first mimeographed and the final revision (see figure 3) in the light of ex- perience included the following inquiries. **In ad- dition to the usual inquiries in regard to name, address, nationality, housing conditions and rent, the items of the schedule center about family com- position; employment, income, surplus and de- ficit; value of property owned; the sickness his- tory of every member of the family during the last twelve months; provision, need and cost of dental work ; the amount and type of life and dis- ability insurance carried by members of the fam- ily ; the dispensary record for all members of the family during the last twelve months; deaths in the family during the year and cost of burial; births in the family during the year ; the employ- ment of physician or mid- wife; and the cost of nursing care and hospital service. A most valu- able part of the schedule was the page assigned to a history of the family known as the * story' in which were entered the significant facts not other- w^ise specifically called for, as, for example, the sickness experience of the family during a period TYPES AND THEIR PROBLEMS 63 of years, their attitude towards physicians, dis- pensaries and social agencies, changes in the standard of living during the year due to sick- ness, complete charity record, etc/ ^ 24 The field staff of this investigation was se- lected on the basis of background of social train- ing, as well as of experience and success in social investigations. Command of languages of the chief immigrant groups of Chicago was another qualification. The field workers were either ad- vanced students of economics and sociology at the University of Chicago and the Chicago School of Civics and Philanthropy, Eegistered Nurses of the Visiting Nurse Association, or experienced salaried investigators. Instruction and supervision of the field work was provided for. Each investigator was given a card of identification by the Commission which certified his appointment as an official agent of the government. Mimeographed instructions were supplied each field worker covering in detail the items on the schedule. Before the field work began conferences of the staff were held for care- ful and detailed interpretation of the schedule and for instruction in regard to methods of approach and technique of investigation. Inexperienced in- vestigators were given a demonstration and su- pervised in the field by the official supervisors of the Commission. Field workers were instructed to state at once to perspective informants the pur- ^*Ihid., p. 181. 64 FIELD WORK AND SOCIAL RESEARCH pose of their visit and the object of the investiga- tion in order to secure intelligent cooperation from the family in making out the schedule. Schedules filled out in the field were promptly edited for conference with the field worker in re- gard to inaccurate or incomplete entries. Out of 2,708 wage-earning families in the ^^ block study/' only 110 gave incomplete reports of total family income, an entry considered one of the most dif- ficult to secure. The method of verifying returns is well out- lined in a paragraph from the report.^^"" *^The Commission availed itself of every opportunity to verify and to correct the data secured by its agents in the field. The leading welfare agencies in Chicago maintain more or less complete records of all the families they serve. Practically all of the more important social agencies register their cases either with the Social Service Eegistration Bureau or with the Central Bureau of the Jewish Charities. Agents for the Commission cleared all the schedules collected in the family study through these two registration bureaus. Through the co- operation of the Cook County Agent all schedules were also cleared through his branch ofiSces in order to verify the statement of the family of the fact of aid received and to determine the exact value in money of the monthly supplies issued. So far as the records made it feasible, all cases were also cleared through the different dispen- 2*a76i