In the 1943, 1947 and 1952 reports containing county level-of-living indexes, indexes for States, major geographic divisions, and the United States were published.

In the 1952 publication indexes for State economic areas were included." These indexes were, in each case, simple arithmetic averages of the indexes of the counties included in the area. In most cases, they differ only slightly from index values that could have been derived by evaluating the formulas for the State, division, or the United States as a whole. For the United States as a whole, the greatest difference for any year between the average computed as the arithmetic mean of the county indexes and the average computed by evaluating the formula for the United States was four index points.

A minor exception to the averaging of county indexes for larger areas arose owing to the problem of combining the small counties. In computing the averages for economic Bogue, Donald J. "State Economic Areas. " U. S. Gov't Print. Off. Washington, D. C. 1951.

areas, States, and divisions, etc., averages of counties and the county combinations were used, with each separate county given a weight of one and each county combination given a weight equal to the number of counties included in the combination. This has only a slight effect on averages for States or larger areas. However, as this method of getting averages was believed to be the best method for 1950, averages for States, divisions, and the United States for years before 1950 were recomputed by this method. Recomputed averages for earlier dates were shown in the 1950 report. For the United States, the slight modification in method of computing the average necessitated a revision for the 1940 index from a previously-published value of 80 to a value of 79. In the case of State averages, the value was changed by one index point in 10 cases, by 2 index points in 2 cases, and by 3 index points in 1 case.

For the 1954 indexes the 1950 index procedure was followed for obtaining the averages for areas other than counties.

Indians on Reservations

The treatment of Indians on reservations has not been uniform in the several Censuses of Agriculture from which the data were taken. In some censuses, an entire reservation was reported as one farm, and in other censuses an attempt was made to obtain a separate schedule for each Indian family operating a farm on the reservation. After consultation with the Agriculture Division of the Bureau of the Census, indexes for certain counties in Arizona, New Mexico, and Utah were not computed for certain years.

Comparability of Indexes For Different Years

In general, the questions from which the data for the level-of-living indexes are obtained were the same or approximately the same in the 1930, 1940, 1945, 1950, and 1954 Censuses of Agriculture, with two minor exceptions. In 1930 and 1954, the number of farms reporting sales of farm products was not published. Therefore, for these two years the fourth item included in the index formula is the average value of sales computed with all farms in the county as the denominator. For other years, the fourth item in the index formula is the average value of sales computed on the basis of farms reporting sales. The second minor noncomparability is with regard to the item of electricity. In 1930, 1940, and 1945, the data related to the number of farms with electricity in the farm dwellings. In 1950 and 1954, data related to the number of farms with electricity. Whereas in these years there may have been a very small number of farms that had electricity in the barn or elsewhere but not in the farm dwelling, the number is believed to be negligible.

RELATED REPORTS

The principal reports that contain previously published level-of-living indexes are: Hagood, Margaret Jarman.

1943. Rural Level-of-Living Indexes for Counties of the United States, 1940. Bur. Agr. Econ. 43 pp. Washington, D. C. (Processed.)

1947. Farm-Operator Family Level-of-Living Indexes for Counties of the United States, 1940 and 1945. Bur. Agr. Econ. 42 pp. Washington, D. C. (Processed.)

1952. Farm-Operator Family Level-of-Living Indexes for Counties of the United States, 1930, 1940, 1945, and 1950. Bur. Agr. Econ. 82 pp. Washington, D. C. (Processed.)

6 In the publication containing the 1950 indexes, indexes were not shown for several counties in these States for which indexes are given in this report. After reexamination of the data it appeared that enumeration in these counties was sufficiently comparable in 1945, 1950, and 1954 to warrant computation of indexes.

Information on construction of the indexes, what they measure, and other related matters is contained in the following articles:

Hagood, Margaret Jarman.

1943. Development of a 1940 Rural-Farm Level-of-Living Index for Counties. Rur. Sociol. 8: 171-180.

1947. Construction of County Indexes for Measuring Change in Level of Living of Farm-Operator Families, 1940-45. Rur. Sociol. 12: 139-150.

Hagood, Margaret Jarman, and Ducoff, Louis J.

1944. What Level-of-Living Indexes Measure. Amer. Sociol. Rev. 9: 78-84.

U. S. GOVERNMENT PRINTING OFFICE: 1957 O 417685