Curves (cd) and (e-f), respectively, exhibit these aver ages. They fluctuate slightly above or below the curve of five year averages, but the increase is evident in each case. They demonstrate that no matter what average is adopted, the resulting curve shows that the absolute1 number of strikes in the United States is increasing. It will be noticed that this method does not carry the "smoothed" curve through the entire period of twenty years. In the case of the five year averages the curve ends with the year 1898. Now, to carry the curve to 1899 it was assumed that the number of strikes for 1901 was at least as large as for 1900 and the years immediately preceding. The number for 1901 was assumed to be the average of 1899 and 1900. This assumption is legitimate, as the number of strikes after 1900 was undoubtedly larger than during and immediately preceding 1900. The reports of the secretary of the American Federation of Labor show a large increase in the case of union strikes after 1900. To extend the curve to 1900 the number of strikes for 1902 was assumed to be at least as large as for 1901. This assumption, again, is unquestionable.® crease. The increase of American strikes can be shown still more accurately. Table II presents the number of establishments affected by strikes for each year from 1881 to 1900. After being "smoothed," as shown in columns (b) the curve (a—b) of Chart III was constructed. It plainly shows a rapid inIt is to be noted that this curve showing the number of establishments affected by strikes is a better standard to judge by, than that showing the simple number of strikes. In the case of the number of strikes, the same prominence is given to a very small strike as to a very large one, while in the case of the number of establishments affected this fallacy is largely obviated. Chart III shows, therefore, not merely the increase in the number of strikes but also the increasing importance of strikes. 'Absolute as distinguished from relative. 2 Repts. Sec. Am. Fed. of Labor, 1901, '02, '03, '04. 3The smoothed figure for 1881 is the actual number and for 1882 is the average of the three years, 1881–2 and 3. U. S. Labor Report, 1901, "Strikes and Lockouts;" also U. S. Labor Bulletin No. 56. Smoothed as above explained. TABLE II.-Establishments affected by strikes in the United States. But the statistics go still farther. Table III shows the number of strikers for each year from 1881 to 1900. It presents the figures both in the absolute and the smoothed form, the smoothed figures consisting of five year averages. Chart IV presents these figures in the form of the curve (a-b). It shows that the number of strikers is increasing. Table III contains also the number of employees affected by strikes each year. When these, in the form of smoothed curves, are plotted, the result is the curve (c-d) of Chart IV. Again, there is an unmistakable increase. Now, it must be noted that these curves of the number of strikers and the number of employes directly affected by strikes are also more important than those of the simple number of strikes, for the same reason as in the case of the number of establishments affected. These curves, not only show the increase in the number of strikes, but the increasing importance of strikes on the side of labor, just as the increase in the number of establishments affected did on the side of capital. |