« SebelumnyaLanjutkan »
Many works have been written relating to statistics but, to the author's knowledge, none have been devoted exclusively to the explanation of the practical part of statistical work.
The first work written in relation to a new branch of a science or art is apt to be doubly unsatisfactory—to the author who is unable to refer to other writers and profit by the criticisms of their work, and to the reader who often fails to find the particular subject in which he is interested given the prominence he thinks it should receive.
The writer of this work had the advantage of ten years' service as an accountant before beginning statistical work. Since June, 1873, or for nearly fifteen years, his duties have been statistical in their nature. Being naturally inventive, , he has not been bound by old methods but has endeavored to introduce improvements in the manner of doing statistical work. He has been advised to put into book form the principal features of his system of practical statistical work, and the results of his experience in the past fifteen years, including his connection with three censuses—two State and one National.
He is aware that his work is tentative; that in a few years there will be much more to write, and much that is
now recommended may be superseded by better forms and improved methods. His aim is to continue to be a student as long as he is a statistician, and he hopes that he may, himself, contribute to future editions of this work, if they should ever be issued, and in other ways, more valuable information than he is now able to supply.
The science of statistics is destined to achieve a great development, and the practical statistician is the one to whom the world looks for honest and assiduous endeavor to broaden the application and elevate the character of the work done in his chosen field of scientific service. The practical statistician must be expert in the use of figures, but such aptitude is only the mere alphabet of his work; he must know the tendencies of the times and select such topics for investigation as will answer the public demand; he must frame inquiries, draft schedules, write instructions for those filling in or answering inquiries, examine the returns, at the same time correcting errors and supplying omissions, adopt and supervise the application of the most improved methods of tabulation, prepare the material for expressive and concise presentation in print, and give therewith such careful explanations and exhaustive analyses that all parties will agree that the truth was his object, that he has found it, and that progressive action may be safely based upon the results presented.
Boston, April 19, 1888.
THE PREPARATION OF SCHEDULES.
Whether statistics are obtained by mail, agents, or enumerators, in nearly all cases it is necessary to prepare forms, blanks, or schedules, upon which to write down the information obtained. Several different forms of work enter into the preparation of schedules.
1. The proper wording of the inquiries. 2. The writing of the general instructions. 3. The writing of the special instructions for each inquiry. 4. Quotations from the law authorizing the collection of the
information. 5. The typographical arrangement of the schedule.
The Proper Wording of the Inquiries. It seems easy to ask questions, and it seems strange that persons receiving schedules by mail do not fill them out properly and promptly and return them when requested. When applied practically it becomes hard to ask even the simplest question and secure uniform and correct
Persons show absolute genius in misunderstanding questions. Some are in a state of honest perplexity and betray it in their attempts to answer and in the letters they often send with their returns.
Of course, we are speaking now of schedules to be filled and returned by mail. A striking instance of statistical mystification was shown by the clothing manufacturer who returned
a manufactures schedule unfilled, with a note saying he was not in want of any goods at that time but when he wished some he would write. Some parties quibble and make as much trouble as possible before answering.
In the preparation of single inquiries, which when combined form the schedule, the intention is to ask a question which will draw out the information desired. In writing, tautology is not a fault if thereby a gain is made in perspicuity. Compound inquiries, or two or more questions in one, should be avoided, for they necessitate compound answers which must be separated by the tabulating clerk. In routine schedules, or those used from year to year, usually by public offices, the best form of inquiry is arrived at after years of experience, and if the officials who answer have long tenures of office the returns are usually correct.
When a person prepares a schedule he should be able to see, in his mind's eye, the results in print. The river can never rise any higher than its source, and the most that any statistical work can contain must be comprehended in the schedule. At the same time the experienced statistician will extract, from an apparently simple schedule by combinations, an array of facts that the casual observer of a schedule would never imagine it contained.
The Writing of the General Instructions. General instructions must be prepared to aid the enumerators or agents who collect the statistics, and also those persons who are called upon to fill schedules and return them by mail.
The general instructions to enumerators or agents should cover such points as correspondence with the main office; directions as to how to send in returns; description of outfit supplied and its use explained ; summary of the work to be done; personal service to be given; a proper day's work; regard for boundaries of districts; the examination and correction of returns before they are sent in ; disposition to be made of materials not used; courtesy to be observed in dealing with the people ; precautions to be taken against false statements by parties supplying information ; the proper course to follow in case of refusal to answer ; concerning the confidential nature of returns; in relation to daily reports of work done ; the keeping of accounts of services and expenses and their return to the office with the necessary certification and vouchers.
The general instructions sent to persons who are to fill schedules and return them by mail should relate to-the definition of the actual time covered by the schedule, or parts thereof, whether the day, month, or year; how to write money values ; description of inercantile values, book values, and combustible values; the use of "symbols" or signs in filling up schedules ; and directions to apply, when necessary, for further instructions by mail.
The Writing of the Special Instructions for each Inquiry. Beyond the filling in of the state, county, town or city, and date, there is seldom an inquiry in a schedule that does not need some special explanation to aid in answering the question. Even when the general and special instructions are supplied, in many cases parties will not take the time or trouble to read them, and a long and expensive correspondence will often be required to gain a few facts which the person could and would have included in the original schedule if he had read and understood his instructions. It is, no doubt, often the case that both inquiries and instructions are lacking in plainness and distinctness, but that is seldom the case with the expert statistician who can see a perfectly filled schedule in his mind, and who knows what to say to secure the answer desired.
Quotations from the Law authorizing the Collection of Information. In the United States the law, like the atmosphere, encircles the citizen, but does not press upon him uncomfortably unless in extreme cases. The American citizen is jealous of his individual rights and opposed, on principle, to inquisitorial inquiries by the government. He is not so much opposed to giving information of a private nature, but he is very solicitous as regards the use to be made of the information. He will give a statistical office individual facts but he wishes, naturally, to be “covered up” in the print. He does not wish his competitors or, it may be, enemies to learn his private affairs and use the information to his detriment. For these reasons, and also to show the individual that “the people” demand the information, a schedule should contain such extracts from the law as will show its general scope and individual application. In many cases the whole law should be printed in the schedule.
The Typographical Arrangement of the Schedule. The typographical arrangement is largely dependent upon the size. The practice in the past has been to use large and unhandy schedules ; the