NOTE PREFATORY TO APPENDIX. The following paper was read before the General Conference of Railroad Commissioners held at Washington, D. C., March 5, 6, 7, 1889, and is inserted as an appendix to this report, because it explains certain statements made in the text, and because it discusses at greater length than was there possible the difficulties encountered in obtaining a torrect railway mileage. It seems appropriate also that in this the first official report on statistics of railways, embracing all the roads in the United States, the extent to which statistical investigation into railway affairs has been developed by the States should be presented. This is done in the table following this paper. The conference before which the paper was read adopted a resolution commending uniformity in the collection and compilation of railway statistics, and it is gratifying to be able to add that since its adjourn ment bills have been introduced into the legislatures of quite a number of the States looking towards the adoption of uniform statistical methods. It is confidently expected that the State commissions and the Federal Commission will soon come to work in complete harmony so far as railway statistics are concerned. UNIFORM RAILWAY STATISTICS. A PAPER READ BEFORE THE GENERAL CONFERENCE OF RAILROAD COMMIS- In the circular of January 31, 1889, inviting the railway commissioners of the various States to participate in a general conference, the subject of uniform railway statistics was given the most prominent place among the topics to be discussed. This probably was by design on the part of those who arranged the programme, and the readiness with which the invitation was accepted may be taken as indicating some degree of interest in this subject. My own experience in the practical work of compiling statistics has not been extended; it has, however, been ample to convince me that without a common understanding on the part of those engaged in this line of investigation, results can not be expected at all commensurate with energy expended. Uniformity of method is a matter of such vital importance to the satisfactory performance of the tasks which government has intrusted to the hands of the several railway commissions, that this convention ought to be unwilling to separate until some definite plan has been adopted for harmonious and co-operative statistical work. The few words I have to say will be directed by this purpose. Permit me then at the outset to call your attention to certain advantages sure to arise from the adoption of uniform methods of collecting and presenting railway statistics in the United States. These advantages are of three sorts. In the first place, uniformity of procedure presents the only feasible plan by which accuracy of conclusions may be insured. In the second place, unless statistics are compiled according to uniform rules it is impossible to use conclusions arising from them for comparative study. And, lastly, it will be very difficult to arrive at complete statistics unless the State commissioners are willing to co-operate with the statistical department of the Interstate Commerce Commission. Let us look at these points a little in detail. The first claim for a uniform rule in statistical inquiry is that it presents the only feasible plan by which accuracy in conclusions may be 344 insured. Consider, by way of illustration, the question of railway mileage. Here, certainly, one would say, is a question of simple measurements, and it is the fault of the statistician if his conclusions are not correct. Without denying the responsibility of the statistician, a moment's consideration will show the question of mileage to be less. simple than it first appears. A great railway corporation is a very complex affair; it is divided into many departments for administration and control, and it is of considerable importance to the statistician which department furnishes the facts to be compiled. Thus the figures on mileage may come from the auditor or. from the engineer, and it is by no means certain that the returns made by the chief engineer will be the same as those made by the financial officer. Discrepancies arising from this cause have so frequently occurred in the experience of our office that they have ceased to be the occasion of surprise. The figures on page 4 in our blank form, for example, which give mileage of road, "included in the revenue account," fail in perhaps one-fifth of the reports received to conform to the figures on page 27, which present details pertaining to the length of line. Here certainly is a discrep ancy, which, unless uniform methods of procedure are adopted by all who compile statistics of mileage, will make agreements in totals impossible. But a more serious source of error presents itself when we notice that the difference between length of line operated and length of line owned is not always kept in view by the officers of roads making reports. Of course the total mileage in the country must equal the sum of the miles of line owned by each and every corporation in the country. Should any company report as mileage owned, mileage which it operates, or controls, there must arise a duplication, vitiating all results. This is very simple and clear when thus presented; and, so far as I am aware, all the reports of the State railway commissions recognize the danger and endeavor to guard against it. But your experience must be very different from that of this office, if you have succeeded in satisfying yourselves that your figures represent things exactly as they are. In the first place, trackage rights present a class of embarrassing questions. It is not possible to be sure in all cases that a strip of road used by several companies is not included in the total returned by each. We have quite accidentally discovered a few cases of erroneous returns of this sort by observing in the expense account an item charged as rentals for trackage rights, although no trackage rights were mentioned in mileage operated. Similar to this is the case of several companies owning jointly a strip of track. One makes return of its portion as part of its mileage owned, while the report of the other road makes no mention of the fact whatever. Another source of embarrassment presents itself |