Geography.  The percentages of recruits from some census regions of the United States have remained fairly stable since the inception of the volunteer force.  However, as Figure 2.10 illustrates, in other regions some substantial shifts have taken place.  The percentage of accessions from the Northeast dropped 8 points from a high of 22 percent in FY 1977 to a low of 14 percent in FY 1989.  Today, 15 percent of enlisted recruits are Northeasterners.  The proportion of accessions from the South increased 9 percentage points from 34 percent in FY 1985 to 43 percent in FY 1995.  In FY 1998, 42 percent of new recruits were from the South.

Changes in geographical representation are related to factors such as shifts in demographic patterns, unemployment, college enrollment, and employment compensation rates, which vary widely across regions of the country.(37)   Obviously, no one factor can explain variations in enlistment rates between different sections of the country; they are more likely attributable to a wide array of economic, social, and demographic factors.

Table 2.10 presents FY 1998 accession statistics by geographic region, division, and state. The third and fourth columns show percentages of accessions and percentages of the 18- to 24-year-old civilian population, respectively, in each area. The fifth column presents military/civilian representation ratios—the percentage of enlisted accessions divided by the percentage of civilians in each area. A representation ratio of 1.00 means that the area has the same proportion of accessions as of the youth population—for example, 8 percent of all recruits and 8 percent of all youth aged 18–24. A ratio of less than 1.00 means that relatively few youth in an area enlist in the military, while a ratio of more than 1.00 indicates above-average market penetration.  The last two columns of the table present the percentages of high-quality accessions (high school graduates in AFQT Categories I–IIIA) and mean AFQT scores for each area.

The South region had the greatest ratio of enlistees (1.2).  The South Atlantic and West South Central divisions had the strongest representation (1.3 each).  The Northeast and North Central regions had representation ratios of 0.8 and the West region had a representation ratio of 1.0.

Slightly more than half of the states had representation ratios of 1.0 or more.  These included:  Maine and New Hampshire in the Northeast; Missouri in the North Central; all states except Utah and California in the West; and all states, except Delaware, Kentucky, Tennessee, and the District of Columbia, in the South.  Among all states and the District of Columbia, the ratios ranged from a low of 0.5 in Massachusetts and Utah to a high of 1.7 in Montana and Wyoming.

The sixth column of Table 2.10 shows the proportion of high-quality accessions by geographical area.  There were only minor differences by region in FY 1998.  The proportion of high-quality accessions by region ranged from a low of 59 percent in the South to a high of 65 percent in the North Central region.  Differences across divisions were somewhat larger.  Nearly 11 percentage points separated the East South Central and West North Central divisions.  Differences at the state level were still larger, ranging from 50 percent in the District of Columbia and Mississippi to 74 percent in North Dakota.

The last column of Table 2.10 shows the mean AFQT score by each geographical area.  Occasionally interest has been expressed in using AFQT scores as an indicator of the performance of state educational systems.  AFQT statistics are not particularly suitable for this purpose for several reasons.  As a sample of youth in a state, ASVAB test-takers reflect a number of selection biases, the total effect of which is unknown.  Those who take the test as part of the enlistment process exclude many students who intend to enroll in college, prospects who fail the enlistment screening test, and youth who do not have an interest in military enlistment.  Therefore, youth who take the ASVAB should not be presumed to be representative of the communities or school systems from which they are drawn.  Even without the biases, it would be difficult to determine how much the test scores reflect differences in school performance from state to state, or how much they reflect other state characteristics, such as social composition and economic conditions.  In sum, while the ASVAB is an excellent instrument for the purposes for which it was designed, it does not provide valid state-by-state school performance data.

Nevertheless, AFQT scores by state may be of interest for purposes other than assessing school system performance.  The AFQT figures in Table 2.10 reflect the mean AFQT percentile scores for accessions in each state.  Percentiles displayed in Table 2.10 are all above 50 because low-scoring applicants are screened out.

Kostiuk, P.F., Geographic Variations in Recruiting Market Conditions (Alexandria, VA:  Center for Naval Analyses, 1989). (go back)

Go to Chapter 3 .