Capt John E. Williams
Lt Col William H. Cummings, III
HQ USAF
Directorate of Military Personnel Policy
Air Force Testing and Survey Policy Team

All services use the Armed Forces Qualification Test (AFQT) composite, a measure of math and verbal skills, derived form the Armed Services Vocational Aptitude Battery (ASVAB) to determine basic aptitude requirements for enlistment (DODD 1304.26). Air Force applicants meeting minimum AFQT requirements are then classified into different specialties according to their scores on four Air Force-unique composites derived from the ASVAB. These composites are Mechanical (MECH), Administrative, General, and Electronic. Even though the Air Force has met accession goals, there have been increasing difficulties in meeting accession requirements for specialties with high MECH entry requirements (J.M. Lynch, personal communication, November 22, 1995). As a result, the Air Force Accession Policy Team asked us to examine possible test-related reasons for these difficulties.

Source: Skinner, M.J., (1996), unpublished data, Armstrong Laboratory

Chart 1 above shows AFQT scores of Air Force accessions increased substantially from CY81-95 while MECH scores steadily declined.

The MECH composite combines scores from three ASVAB subtests: General Science (GS), Mechanical Comprehension (MC), and Auto and Shop Information (AS, double-weighted).

Source: Skinner, M.J., (1996), unpublished data, Armstrong Laboratory

Chart 2 above shows a breakdown of MECH into its component subtests for Air Force accessions. There were small changes in GS and MC scores, but a clear drop in AS scores from CY81-95.

These score trends in Air Force accessions raise three questions: (1) Were the trends caused by some Air Force accession policy or practice, or are they characteristic of applicants in general? (2) Did the same trends hold for other service applicants or are they Air Force unique?, and (3) To what can the trends be attributed?

To help answer these questions, we obtained applicant data for all four services from the Defense Manpower Data Center (DMDC). These data gave us a "big picture" perspective of everyone who applied for military service.

Source: Derived from unpublished data provided by Jordan, L.A. (1996), DMDC

Chart 3 above shows Air Force applicants, like Air Force accessions, experienced a decline in MECH scores and a steady increase in AFQT.

Source: Derived from unpublished data provided by Jordan, L.A. (1996), DMDC

Chart 4 above shows Air Force applicants, like Air Force accessions, experienced a clear drop in AS scores and little change in GS and MC. This finding suggests that trends seen in Air Force accessions reflect trends in Air Force applicants and are not a result of accession policies or practices.

Source: Derived from unpublished data provided by Jordan, L.A. (1996), DMDC

Chart 5 above shows applicants for all services combined. While there is some variance between services, all follow similar trends. Overall, there is a clear drop in AS scores (particularly beginning in FY83) and smaller changes in GS and MC. The MECH subtest trends hold for the entire military applicant pool, not just Air Force applicants.

We examined five reasons for the trends seen in Air Force applicants:

1. Ethnic Group: Did an increase in minority applicants cause the drop in AS scores? Answer: No. The relative proportions of white, black, and others for the Air Force (and other services) remained relatively steady for FY81-95. AS scores also decreased for all ethnic groups from FY 81-95.

2. Quest for High Aptitude: Has the Air Force’s quest for ever-higher AFQT scores hurt our ability to access folks with MECH skills? Answer: No. As shown in Table 1 below, AFQT and MECH for individuals are positively, not negatively, correlated.

CORRELATIONS OF THE MECH COMPOSITE WITH THE AFQT

Source: Sawin, L., (1996), unpublished data, Armstrong Laboratory

3. Sex: Did an increase in female applicants contribute to a drop in MECH scores? Answer: Yes, probably accounting for part of the drop.

Source: Derived from unpublished data provided by Jordan, L.A. (1996), DMDC

Chart 6 above shows all services (except Marines) had increases in the percentage of female applicants. Average MECH scores for Air Force female applicants are consistently lower than male applicants—averaging 27.2 percentile points lower from FY81-95. However, this difference does not account for all the drop in MECH (or AS): Male applicants’ MECH (and AS) scores also decreased for FY81-95.

4. National Trends: Are there fewer people with mechanical skills in the applicant pool? Answer: Quite likely. As previously noted, AS scores decreased for all services’ applicants, even though AFQT scores rose. For Air Force, Navy, and Marines (and to a lesser extent, Army), the decrease in AS scores was strongly correlated with the passage of time. This hypothesis is also supported by reports of employers unable to fill positions for qualified mechanics (Orenstein, 1995). Concerns have also been voiced by the Society of Automotive Engineers over shortages of mechanics and difficulties in recruiting auto mechanics (Eldridge, 1996).

5. Test Obsolescence: Is increasing obsolescence of the AS subtest contributing to the decline in AS scores? Answer: Possibly.

Members of this office and the Air Force Accession Policy Team recently reviewed the AS subtest on the paper and pencil ASVAB 18/19. The AS subtest from each test version contained several items on non-current automotive technology (e.g., carburetor-type fuel systems, non-R-34 air conditioning systems, and non-electronic ignition systems).

In early October, 1996, we raised our concerns to the Manpower Accession Policy Working Group (MAPWG), which oversees the ASVAB program. Several members of the MAPWG believe the AS subtest continues to be valid. However, DMDC has agreed to provide us with item statistics for review. After we receive these statistics, if they indicate item obsolescence, we will work through the MAPWG to correct the problem.

1. Increasing AFQT scores and decreasing MECH scores of Air Force accessions match trends seen in Air Force applicants and other service applicants. They do not appear to result from Air Force accession policies or practices.
2. Decreasing Air Force MECH composite scores are most likely due to:

1. An increase in the percentage of female applicants (who score substantially lower than male applicants on the MECH composite).
2. A general decline in AS scores in the applicant pool.

3. The general decline in AS scores may be a result of:

1. A decline in knowledge of automobiles, tools, and shop terminology and practices in the applicant pool.
2. Increasing obsolescence of the AS subtest.

We would like to thank Dr. Jacobina Skinner, Dr. Linda Sawin, and Ms. Linda Jordan for their assistance with this report.

DODD 1304.26, Qualification Standards for Enlistment, Appointment, and Induction, December 21, 1993.

Eldridge, E. (1996, March 4). Carmakers prepare for auto recycling. USA Today, p. 7B

Orenstein, B.W. (1995, November 20). Machinist shortage plagues manufacturers. Eastern Pennsylvania Business Journal.