ASVAB to IQ at a glance

An AFQT percentile of 75 points to an IQ of about 108, with a likely range of 97–120. A 50 points to about 100, and a 93 (the bottom of Category I) to about 118.

Two things set the AFQT apart. It is a percentile against a general-population sample, so there is no guess about who takes the test. And it is dominated by vocabulary and school maths, so it tracks IQ well in the general population but says less about fluid reasoning.

Notable ASVAB scores and the IQ they point to
ASVAB score	Percentile among test-takers	Estimated IQ	Likely range
10	10th	85	73 to 96
21	21th	90	79 to 102
31	31th	94	83 to 106
50	50th	100	88 to 112
65	65th	105	93 to 116
75	75th	108	97 to 120
85	85th	112	101 to 124
93	93th	118	106 to 129
95	95th	120	108 to 131
99	99th	128	116 to 139

ASVAB to IQ conversion chart

Every row below comes from the same model as the calculator. The likely range covers about four in five people at that score. The last column shows what the IQ would be if the ASVAB were a perfect measure of IQ, which is the number many other converters publish.

ASVAB to IQ chart (AFQT percentile 1 to 99, calculated from four ASVAB subtests)
ASVAB score	Test-taker percentile	Estimated IQ	Likely range	If a perfect measure
10	10	85	73 to 96	81
15	15	88	76 to 99	84
20	20	90	78 to 101	87
25	25	92	80 to 103	90
30	30	94	82 to 105	92
35	35	95	84 to 107	94
40	40	97	85 to 108	96
45	45	98	87 to 110	98
50	50	100	88 to 112	100
55	55	102	90 to 113	102
60	60	103	92 to 115	104
65	65	105	93 to 116	106
70	70	106	95 to 118	108
75	75	108	97 to 120	110
80	80	110	99 to 122	113
85	85	112	101 to 124	116
90	90	115	104 to 127	119
95	95	120	108 to 131	125

IQ to ASVAB: what score to expect

Going the other way, this table shows the ASVAB score people at each IQ typically get. Scores at the same IQ vary widely, so the range matters more than the middle figure.

Typical ASVAB score by IQ
IQ	Typical ASVAB score	Likely range	How rare is this IQ
80	14	3 to 38	1 in 11 at or below
85	21	6 to 49	1 in 6 at or below
90	30	10 to 59	1 in 4 at or below
100	50	22 to 78	1 in 2 at or above
110	70	41 to 90	1 in 4 at or above
115	79	51 to 94	1 in 6 at or above
120	86	62 to 97	1 in 11 at or above
125	91	71 to 98	1 in 21 at or above
130	95	80 to 99	1 in 44 at or above
135	97	86 to 99	1 in 102 at or above

One IQ, several tests: ASVAB next to other scores

The same IQ shows up as very different numbers on different tests. This table shows the typical score at each IQ on ASVAB and on the tests closest to it. Read across a row to compare, for example, what an SAT and an ACT would each be for the same person.

Typical score by IQ across ASVAB, Wonderlic, CCAT, SAT, ACT
IQ	ASVAB	Wonderlic	CCAT	SAT	ACT
100	50	21	23	990	17
110	70	25	27	1140	21
120	86	28	32	1280	25
130	95	32	36	1410	30
140	98	36	40	1530	33

These are typical scores, not guarantees. Every test has its own likely range, shown on its page.

AFQT, ASVAB and which number to enter

The ASVAB is a battery of subtests. The AFQT is a single percentile derived from four of them: Word Knowledge and Paragraph Comprehension (combined as verbal expression), Arithmetic Reasoning and Mathematics Knowledge. The formula counts verbal twice, as twice verbal expression plus arithmetic reasoning plus mathematics knowledge.

Most people mean the AFQT when they say "my ASVAB score". It is the number that decides whether you qualify to enlist. The line scores that decide which jobs you can do are different numbers and do not go into this converter.

Why this conversion is the most defensible one

Almost every other test here is taken by a self-selected group, so converting a score to IQ means first guessing how able that group is. The AFQT percentile is calculated against a nationally representative sample of young Americans, not against enlistees. A 75 means you scored higher than about three in four of that whole cohort.

That turns the conversion into a standard percentile-to-IQ calculation, with a correlation added on top. It still has two weaknesses. The reference sample was tested in 1997, so it is nearly three decades old and the Flynn effect has moved things slightly since. And the ASVAB is heavily loaded with learned vocabulary and school maths, so it relates to crystallised knowledge more than to fluid reasoning.

What research says about the ASVAB and IQ

The ASVAB's first factor is a general-ability factor, and researchers have used it as a stand-in for IQ. The SAT and ACT studies cited elsewhere on this site did exactly that: Frey and Detterman (2004) and Koenig and colleagues (2008) took their measure of general intelligence from ASVAB subtests. An older comparison put the AFQT and the WAIS at about .80, though that figure is from the 1970s and we could only confirm it through secondary sources.

A 2000 paper by Roberts and colleagues argued that the ASVAB is little more than acculturated learning, meaning crystallised knowledge. That fits the subtests, which are vocabulary, reading and school maths. It does not make the ASVAB a poor measure of general ability in the population, only a narrower one than a matrix test.

Mensa does not accept the ASVAB as a qualifying test.

Using an AFQT-to-IQ estimate

The AFQT decides eligibility to enlist, and line scores decide which jobs you can train for. An IQ estimate does neither. It is a way of translating your percentile into the scale most people mean by IQ, with an honest range, and of comparing it with other tests.

If you scored in a category below the one you were hoping for, note that the AFQT is retakeable under service rules, and that vocabulary and school maths respond to study. A retest gain says something about preparation, not about a change in ability, and it is worth keeping the two apart when you read your own estimate.

What your AFQT means

The categories are the military's own. The estimated IQ column shows the most likely value for the percentile at the ends of each band.

What ASVAB score bands mean
Score	Estimated IQ	What it means
Category V (1 to 9)	72 to 84	Below the minimum for enlistment in all services.
Categories IVA to IVC (10 to 30)	85 to 94	Below the middle. Enlistment depends on the service, the year and other qualifications, and often requires waivers or programmes.
Category IIIB (31 to 49)	94 to 100	Just below the national median. Enlistment minimums vary by service and year, so check current requirements.
Category IIIA (50 to 64)	100 to 104	Around the median to somewhat above.
Category II (65 to 92)	105 to 117	Above average. A large group, and the range that qualifies for most technical roles.
Category I (93 to 99)	118 to 128	The top 7%. Qualifies for the widest set of jobs, including the most selective programmes.

Why a ASVAB score is not an IQ

It leans on learned knowledgeVocabulary, reading and school maths dominate the AFQT. Someone with strong reasoning but weak schooling scores lower than their ability.
The norms are from 1997The reference group was tested nearly 30 years ago. Population changes since then are small but real.
The percentile is coarse at the topA 96, a 98 and a 99 look very different but mean small differences in raw performance, and the estimate barely moves.
Retesting and preparation matterStudying vocabulary and maths raises the score, so it is partly a measure of preparation.

The evidence behind these numbers

Evidence grade: Moderate

The percentile is general-population referenced, and the ASVAB's general factor has been used as an IQ measure for decades. The AFQT-to-Wechsler correlation is old and confirmed only through secondary sources.

Published facts we used

AFQT percentile is referenced to a nationally representative 1997 sample of 18 to 23 year olds (official ASVAB site).
The AFQT and the WAIS correlate about .80 (McGrevy, Knouse and Thompson, 1974), as reported in secondary sources.
The ASVAB's first factor has been used as the general-ability criterion in the SAT and ACT validation studies.
The ASVAB is dominated by crystallised abilities (Roberts et al., 2000).

Assumptions we made

Mean IQ of the reference group: 100, by construction of a general-population sample.
IQ spread: SD 15.
Correlation between AFQT and IQ in the population: .80, taken from the older AFQT and WAIS comparison.

Assumed test-taker IQ: mean 100, SD 15
Assumed correlation (r): 0.8
Likely range: The score at which about 80% of people with that estimate fall, from the residual spread after regression
Norms: AFQT percentile referenced to the 1997 nationally representative youth sample.

ASVAB to IQ: frequently asked questions

What AFQT score equals an IQ of 100?+

An AFQT percentile of 50 corresponds to an estimated IQ of about 100, since the AFQT is scored against the national average. People with an IQ of 100 typically score around 50, with a range of 22–78.

What is the AFQT score for an IQ of 130?+

People with an IQ of 130 typically score around 95 on the AFQT percentile scale, with about four in five between 80–99.

Is the ASVAB an IQ test?+

No, but its general factor correlates strongly with IQ and researchers use it as a measure of general ability. It leans on vocabulary and school maths, so it captures crystallised more than fluid intelligence.

What is a good ASVAB score?+

It depends on your goal. Each service sets a minimum AFQT for enlistment, and higher line scores open more jobs. An AFQT of 93 or above is Category I, the top group.

Does Mensa accept the ASVAB?+

No. US Mensa's published qualifying table states the ASVAB is not accepted.

Should I enter my ASVAB line score?+

No. Enter your AFQT percentile, the single 1 to 99 number reported on your ASVAB results. Line scores use different scales and would give a wrong answer.

References & Sources

Official ASVAB (2026). ASVAB scores: AFQT and standard scores. officialasvab.com. https://www.officialasvab.com/recruiters/scores/
Roberts, R. D., Goff, G. N., Anjoul, F., Kyllonen, P. C., Pallier, G., & Stankov, L. (2000). The Armed Services Vocational Aptitude Battery (ASVAB): Little more than acculturated learning (Gc)?. Learning and Individual Differences, 12, 81-103. http://www.iapsych.com/iqmr/roberts2000.pdf
Frey, M. C., & Detterman, D. K. (2004). Scholastic assessment or g?. Psychological Science, 15(6), 373-378. https://journals.sagepub.com/doi/abs/10.1111/j.0956-7976.2004.00687.x
American Mensa (2026). Qualifying test scores. us.mensa.org. https://www.us.mensa.org/join/testscores/qualifying-test-scores/