Air Force Line Scores: MAGE Composites Explained

Air Force job qualification runs on four composite scores called line scores, not on the AFQT. An AFQT in the 70s is no guarantee that a given career field is open, and the conversation at MEPS that closes one off usually turns on a line score rather than on a medical issue or a quota.

The AFQT decides whether you can enlist. The line scores decide which jobs you can hold. They are built from a different mixture of subtests, they are reported separately, and they are the reason the technical half of the ASVAB matters even though it contributes nothing to the score everyone talks about.

If you have a particular Air Force job in mind, this is the page that connects your test to it. If you do not have one yet, read this and then go browse the AFSC library, because it is far easier to study toward a target than toward a vague hope of doing well.

The short version:

• The Air Force uses four composites: Mechanical, Administrative, General, Electrical, together called MAGE
• Every enlisted AFSC sets a minimum on one or more of them
• They are percentiles from 0 to 99, not percentages, so a 60 is good
• The composites overlap, which makes some subtests worth several times more than others
• Arithmetic Reasoning and Mathematics Knowledge feed two composites each plus the AFQT
• Assembling Objects feeds nothing in the Air Force
• The Air Force does not publish the arithmetic weightings, so treat every contributing subtest as important

What MAGE actually is

Four numbers make up the system.

Those descriptions are tendencies, not rules. Plenty of careers set minimums in two areas, and a few set them in three. What matters is that each AFSC publishes its own requirement, and your four line scores either clear it or they do not.

Other branches do this differently. The Army uses ten composites, the Navy uses its own set. The MAGE structure is Air Force specific, which is why generic ASVAB advice about "your line scores" is often about somebody else's test results.

Which subtests feed each composite

Here is the map. Read it as which subtests contribute, not as a formula.

Three of the four are settled. Mechanical is not, and the third column is not hedging for its own sake: published versions of the Mechanical formula conflict. One widely reprinted version gives General Science, Mechanical Comprehension and a doubled Auto & Shop. Another gives Arithmetic Reasoning, a doubled Verbal Expression, Mechanical Comprehension and Auto & Shop. A third drops the doubling. The older of these also states an Administrative formula built from Numerical Operations and Coding Speed, two subtests the Department of Defense removed from the ASVAB in 2002 and 2003 - which is a good reason to distrust its Mechanical figure as well.

What every version agrees on is that Mechanical Comprehension and Auto & Shop Information are in it. Those two are also the only subtests that feed nothing else, so for anyone targeting a Mechanical minimum they are the work that cannot be done anywhere else on the test.

The fourth composite goes by two names. Air Force requirement data labels the area Electrical, which is the spelling used throughout this site and in the career tables, while most published ASVAB material calls it Electronics. They are the same score.

A note on Verbal Expression, since it appears twice. VE is not a subtest you take. It is Word Knowledge and Paragraph Comprehension combined into a single verbal score, and it is the same VE that gets doubled inside the AFQT formula. So when the table says Administrative draws on Verbal Expression, it means your Word Knowledge and Paragraph Comprehension results, together.

Flip the table around and you can see, subtest by subtest, where your work lands:

That grid is the most useful thing on this page. Everything below is a consequence of it.

Why we do not print the weightings

Other sites publish the composite formulas with exact coefficients, complete with which subtests get multiplied by two. This page does not, for a specific reason.

The Air Force does not publish the arithmetic weightings of its composites. Published third-party versions of those formulas disagree with each other, and none can be verified against an authoritative source. They also disagree about which subtests are in the Mechanical composite at all, and at least one widely reprinted version still lists subtests that were removed from the ASVAB over twenty years ago. Printing a coefficient from that material would be worse than printing nothing, because somebody would plan their studying around it.

There is a further reason the raw formula would not help you. The numbers that go into a composite are not your percent correct; they are scaled scores, and the result is converted to a percentile against a reference population using tables the Department of Defense does not publish either. Even a correct formula would not let you compute your own line score at home.

What is well established, and what we have given you above, is which subtests contribute to each composite. It is also well established that the contributions are not equal - some subtests carry more weight inside a composite than others.

So the practical guidance is simple and it does not require the formula:

Treat every contributing subtest as if it matters, because you cannot know which one is doing the heavy lifting. That is a more robust plan than optimizing against an unverified coefficient, and it is the same plan the real numbers would produce.

Line scores are percentile-style, not percentages

Line scores work the way the AFQT does on this point, and it is worth being explicit.

A line score of 60 does not mean you answered 60 percent of anything correctly. Unlike the other branches, which add standard scores together, the Air Force reports its composites as percentiles from 0 to 99 - the same kind of number as your AFQT. A 60 means you outscored 60 percent of the reference group.

This is why a job requiring a General score of 55 is a real requirement and not a formality, and also why the gap between a 47 and a 62 is much larger in practice than fifteen points sounds. Read your score sheet knowing that the numbers are compressed: most people are bunched in the middle, and moving from average to good takes fewer additional correct answers than the spread of the numbers suggests.

For the same reason, do not dismiss a job because its minimum sits a few points above your practice score. A modest improvement in the right subtest moves a composite a long way.

The overlap is the whole strategy

Look at the grid again and count. Some subtests appear in two composites. Some appear in one. One appears in none. And four of them also feed the AFQT.

That means subtests are not worth the same amount of your study time, and the difference is enormous.

Read the top two rows carefully. Arithmetic Reasoning and Mathematics Knowledge each move two composites and the AFQT, and unlike the verbal pair they do it without sharing credit with another subtest. If you have one weekend to study, it goes here. There is no close second.

The verbal pair is close behind but works differently. Word Knowledge and Paragraph Comprehension combine into VE, so each is roughly half of a score that feeds two composites and gets doubled inside the AFQT. Individually they matter slightly less than AR or MK; together they matter more than either.

General Science is the one most often skipped. It contributes nothing to the AFQT, which is why it gets skipped, and it feeds two composites, which is what makes skipping it expensive. If your target job gates on Mechanical or Electrical, General Science is not optional.

Assembling Objects is worth nothing to you

Assembling Objects does not count toward the AFQT and does not feed any of the four Air Force composites. Zero of five scores.

Other branches use it, so it stays on the test and you will take it. But study hours spent on spatial puzzles are hours taken from Arithmetic Reasoning, where they would have moved three numbers.

If you want to check that for yourself, the Assembling Objects guide explains what the subtest asks and why it lands nowhere on an Air Force score sheet.

Answer the questions, do not leave blanks, and give it none of your preparation time. That is the complete strategy for that subtest.

What a job's requirement looks like

An AFSC requirement is written as a minimum in one or more aptitude areas. In practice it reads like this:

When a job lists two areas, you need both. Clearing one and missing the other by a single point closes the job, and that is a common way for an otherwise strong applicant to be turned down.

Minimums across the enlisted force run from the high 30s to the mid 60s. The technical and analytical fields cluster at the top of that range; the broadest-entry fields sit lower. Nothing is uniform, which is exactly why a generic score target is useless and a specific one is not.

Beyond the line score, many careers add their own requirements: security clearance eligibility, vision or color vision standards, physical demands, citizenship, a shredout selection after tech school. A line score gets you considered, not assigned.

Work backwards from the job, not forwards from the score

The applicants who end up satisfied almost all did the same thing in the same order.

1. Pick two or three career fields you would actually be happy in. Not one - availability changes, and a single-job plan is a plan to be disappointed.
2. Look up the composite minimums for each. Every page in our AFSC library lists them.
3. Note which composites appear. Usually you will find your shortlist gates on one or two areas, and that tells you which subtests to study.
4. Take a full practice test and score it honestly. You need a baseline in the same units as the requirement.
5. Study the gap. Not the whole ASVAB. The subtests feeding the composites your shortlist needs, plus whatever the AFQT still requires.

The order matters. Studying everything equally is how people end up with a good AFQT and mediocre composites. A test plan built from a job list beats a test plan built from a study guide's table of contents every time.

And if your shortlist turns out to gate on Mechanical while your strengths are verbal, better to learn that now, with months to work, than at MEPS with a recruiter waiting.

AFQT and line scores are two separate gates

The two gates are separate, and collapsing them into one idea leads to bad study decisions. Side by side:

You must clear both. Clearing one tells you nothing about the other. For the full explanation of the first column, see AFQT score explained.

How to spend your study time

Assume limited hours, because everyone has limited hours. This ordering holds for most applicants:

1. Get the AFQT safely clear of the floor first. If you cannot enlist, nothing else on this page applies. That means the four AFQT subtests, with the verbal pair first because VE is doubled.
2. Then the two math subtests, if they are not already strong. They are already on your list from step one, and they also count toward the most composites. Nothing else pays twice like this.
3. Then General Science. Two composites, low effort, widely skipped.
4. Then the one technical subtest your target composite needs. Electronics Information for Electrical jobs. Mechanical Comprehension and Auto & Shop for Mechanical jobs.
5. Assembling Objects: nothing. Show up, answer, move on.

One rule applies to every subtest on the test. Answer every question, including the ones you are guessing on and including Assembling Objects. There is no penalty for a wrong answer on the ASVAB, and on the computer-adaptive version an unanswered question at the end of a section can cost you more than a wrong one.

Now go pick a job. The AFSC library lists all 170 enlisted career fields with the composite minimums each one demands, and that number is the only score target worth studying toward.