Publications of the American Statistical Association Volume 10 (Paperback)

This historic book may have numerous typos and missing text. Purchasers can download a free scanned copy of the original book (without typos) from the publisher. Not indexed. Not illustrated. 1908 Excerpt: ...in the Dewey Report taken from each section. TABLE III. The Distribution Of Adult Male Waoe-earners Bt Geographic Divisions. From this table it appears that, geographically considered, Professor Dewey's selection of wage-earners is only fairly representative. The Middle States and the West, as a whole, are given proper weight, but New England and the Southern States are somewhat neglected, while the Central States are noticeably over-represented. The question immediately arises, has this defect in the selection of data appreciably influenced the results of the investigation? This question may be answered by a simple test. The Census of Manufactures provides us average wages in different sections of the United States, as follows: --1890. 1900. New England States $509.33 $507.12 Middle States 540.62 528.71 Southern States 364.83 334.96 Central States 4704 488.51 Western States 565.91 577.09 Pacific States 620.39 577.11 These averages, though inaccurate for some purposes, are sufficiently accurate to show whether a general average based upon them and weighted by the actual proportions of wage-earners in the several sections of the country would differ materially from a similar general average weighted by the proportions which characterized Professor Dewey's selection. The results of this test (Table IV) show that in 1890 the average weighted according to the true proportions differs from the average weighted according to Professor Dewey's proportions by only ten cents. In 1900, however, the over-weighting of the Central Division is shown to have exercised an appreciable influence. It appears above that the only important manufacturing sectioD of the country showing an increase is the Central States. In short, from these results it seems fair to conclude that Pr...