AFQT scores are the primary measure of recruit potential. Figure 2.7 indicates the percentage of NPS recruits who scored at or above the 50th percentile (Categories I–IIIA) since FY 1973. Numerical data are in Appendix D, Table D-12. The drop in Category I–IIIA recruits after FY 1976 was due primarily to the miscalibration of the ASVAB.  In FY 1976, when new versions of the ASVAB were introduced, an error in calibrating the score scales made the new versions "easier" than the old versions (i.e., applicants received test scores higher than their actual ability). In FY 1980, an independent study of the calibration was made and the test was correctly calibrated. Then, Congress added legal provisions stipulating that no more than 20 percent of accessions could be in Category IV and that such accessions had to be high school diploma graduates.  However, as previously stated, Defense Planning Guidance decreases this limit even further, allowing no more than 4 percent of recruits to come from Category IV.
One ASVAB sub-measure, the Armed Forces Qualification Test (AFQT)--which is just a separate analysis of the same tests that focuses only on four of the subtests--is also population-referenced, but its score from 1-99 is a percentile score. This means that if you score a 73, you scored as well as or higher than 73% of test-takers in the national sample.