6. The fundamental cause for the high energy of the pioneer was the action of expulsive, attractive and winnowing selection in the migratory process.

7. Recent interstate and rural-urban migrations have also apparently tended strongly to select the energetic and intelligent. 8. The characteristic of being energetic tends to be transmitted by heredity.

9. The character of the American people as a whole is strikingly parallel to that of the pioneer.

10. The type of individual left behind in interstate migrations, and the extreme type of ultra-pioneer, were neither of them as desirable as parents as were the pioneer settlers.

11. The welfare of the coming generation can best be promoted by a greater endowment of energy under better control.

IV. SELECTIVE MIGRATION IN IOWA

In 1850 Iowa was on the frontier and was being settled by the pioneers whose characteristics have been discussed in previous chapters. In 1910 Iowa was reported as the only state in the United States which had lost population during the preceding ten years. The analysis of interdivisional migration (see page 36) showed that the West North Central States, of which Iowa is one, have been among the heaviest losers in recent decades through rural emigration. Iowa, therefore, is a particularly appropriate state to study for details as to the operation and the effects of the recent type of migration. The fact that Iowa has a state census at dates midway between the dates of the federal censuses facilitates such an inquiry.

16. THE MEASUREMENT OF RECENT MIGRATIONS IN IOWA

The application to Iowa data of the age-distribution method of measuring migrations involves certain special problems. The first of these is due to the fact that Iowa is not a part of the Registration Area, and does not have complete registration of deaths. This makes necessary the estimation of survival rates by methods less direct and accurate than those used for the United States as a whole. The second problem lies in the use of state census data in conjunction with federal census reports for the state, so as to reduce the intercensal periods to five instead of ten years. The state censuses report age distribution according to different age groupings than those used by the federal census, so that extensive interpolations must be made. Also the question arises whether the state censuses are of the same degree of accuracy as the federal censuses. These special problems will be discussed in detail in the next few pages.

DEATH RATES

The incompleteness of the death reports published for Iowa by the state authorities is presumable in view of the fact that the government has not admitted the state to the registration area; but the number of deaths reported may be used as a lower limit, above

which the true number of deaths must lie. Moreover since the variations in the death rates from one community to another are determined chiefly by the age and color distributions of the respective populations and the proportions of the respective populations living in cities, it is possible to make fairly accurate estimates of the deaths by ages in Iowa by weighting according to Iowa conditions the death rates by ages in the cities and the rural districts of the registration area in 1910. No correction for chronological changes has been attempted. In as much as Iowa is a state where the high economic level of the population insures a relatively low death rate, the estimate based on conditions in the registration area may safely be taken as a maximum rate for the state. A third factor which reduces the error from incorrect death rates is the fact that the death rates are lowest at the ages in which migration is most frequent. The force of these three factors is more readily appreciated by examining Table 23 and Chart 8 in

TABLE 23

IOWA DEATH RATES

per 1000 by ages, as calculated from state reports for 1915 and as estimated from rural and urban death rates in the original registration states in 1910. Based on data from 106; 111: pp. xix, lxxiv. See also 151.

Ages

0-1 1-5 5-10 10-20 20-30|30-40 | 40-50 56.1 4.6 1.5 1.7 3.2 4.3 6.4 108.4 11.0 2.9 2.9 5.3 6.7 50-60 60-70 70-80 80-90 11.5 26.7 66.6 132.7 18.1 36.2 82.0 174.9

9.4

90

All ages

192.2

9.5

334.9

14.5*

*Weighted according to age distribution of Iowa population.

which the death rates by ages in Iowa in 1915, as estimated from state reports, are compared with the death rates by ages estimated by weighting the death rates by ages in the original registration states for males and females in cities and in rural districts according to the importance of corresponding groups in Iowa.

The effect of this difference in estimated death rates upon the estimate of migration may be illustrated by taking the period of life in which migration is most frequent-10 to 30 years of age. In 1900 there were 864,627 persons of these ages returned for Iowa by the federal census (125 p. 374). If the death rates estimated from the registration area were operative in the state,

IOWA DEATH RATES BY AGES.

as calculated from data reported for 1915 by state officials and as estimated from death rates for rural and urban districts of the original registration states in 1910. (See Table 23.)

approximately 822,500 of these persons would survive in 1910 (at which time their ages would of course, range from 20 to 40 years). The 1910 census returned only 703,043 persons 20 to 40 years of age in Iowa. (125 p. 374) This would leave 822,500-703,043, or approximately 119,500 persons who presumably must have left the state during the years 1900 to 1910 in excess of the number of these ages who entered the state during the same period. This would amount to a net emigration of 14.5 percent of the survivors. If on the other hand the state reports were correct, and the low death rates which they show were actually in operation the result would indicate that 837,500 persons survived instead of the 822,500 indicated by the registration area rates. The net emigration in this case would be approximately 134,500 instead of 119,500, or 16.1 percent instead of 14.5 percent of the survivors. These estimates, as indicated above, may be assumed to be the upper and lower limits between which the correct number lies.

The probable percentage of error in estimates of emigration for the older age groups would tend to be larger, though the absolute error in number of emigrants would be much smaller. Of all persons five years of age or over in the state in 1900, the net emigration during the decade 1900-1910 was 232,600 if the registration area death rates are used, and 282,500 if the state report rates are used. If, therefore, the only source of error were the death rates, we might say with considerable finality that between 230,000 and 280,000 persons left the state in excess of those entering it between 1900 and 1910.

STATE VS. FEDERAL CENSUS TOTALS

The

Two other sources of error must, however, be examined. first of these is suggested by the contention of the state census authorities that the 1900 census returns for Iowa were 100,000 too large and the 1910 returns 75,000 too small. (111: p. xviii). This contention is made, as the report states, "without any pretense of information on which to base the assumption." The object avowed is to secure an estimate of population increase "more in harmony with known facts" and to show a steady increase in population instead of irregular gains and losses from 1900 to 1915.

If these assumptions of the state authorities were correct, the amount of emigration from the state during the decade 1900-1910 would be less than half the estimate given above, while the migration between 1910 and 1915, instead of being a gain of 8,000 would be a loss of 80,000 or so. The best test as to accuracy of the federal and state census returns would be a comparison with independent sources of information. Probably the most satisfactory

TABLE 24

FEDERAL, STATE AND SCHOOL CENSUS RETURNS COMPARED

Number of persons five to 20 years of age in Iowa as reported by the school, state and federal census, 1890-1919. Sources of data 110: p. xi; 109; 111: p. lvix; 121: p. 215: 125: p. 374: 112.