AFOQT Scores and Results: What They Mean

Your result is not one number

The AFOQT does not produce a single score. It produces a set of composites — scores built by combining performance across several subtests — and different composites matter to different career fields. A candidate applying for a flying slot and one applying non-rated are effectively being measured on different things by the same exam.

This is the most important structural fact about AFOQT scoring, and it has a direct planning consequence: before you decide what to study, you need to know which composites your career field actually requires. An hour spent on a subtest that feeds nothing you need is an hour wasted, however diligently spent.

Composites are percentiles, not marks

Composites are reported on a percentile scale. A score of 70 does not mean you answered 70 percent of anything correctly — it means you performed better than roughly 70 percent of the reference group the test is normed against.

Two things follow from that, and both matter:

• The difficulty of your particular form does not disadvantage you. The scale already accounts for it. A harder set of questions does not produce a lower percentile if everyone found it harder.
• You are being compared to other officer candidates, not to the general population. That reference group is largely degree-holding and self-selected, which is why performance that looked strong on a broader test can land mid-pack here. It is not that you got worse; the comparison group changed.

Two figures we can now give you, and one we still cannot

Until August 2026 this section withheld all three, because the governing document — DAFMAN 36-2664, Personnel Assessment Program (17 January 2025, incorporating Change 1, certified current 15 July 2025) — was recorded in our register as unverified. We have now read it, which resolved two of them and settled the third in a way worth stating plainly.

Your best composite counts, not your most recent. Superscoring is applied across all composites: your highest score on any attempt becomes the score of record. A later, weaker sitting does not displace a composite you have already earned. (A2.7) The retake page works through what that means for the decision to sit again.

Answer sheets are scored centrally, and results are posted within a week of arriving. The manual states that results appear on the testing office website within one week of the answer sheets reaching it from the testing facility. (A2.5) Note what that clock is measured from — the sheets arriving at the scoring office, not you finishing the exam — so the wait you experience includes however long the despatch takes.

What our practice reports can tell you

After a test here you get accuracy overall and by subtest, a pacing breakdown showing where the clock beat you, a count of unanswered items, and your weakest areas in order. That is genuinely useful: it is exactly the information needed to decide what to work on next.

This is your performance on our original practice material. It is not an AFOQT score, percentile or composite. The Air Force does not publish the conversion tables that would make such a calculation possible, so nobody outside the Air Force can produce one.

We use the terms practice performance and diagnostic estimate deliberately, and never describe our output as an AFOQT score, percentile or composite. Where a figure rests on a handful of questions we say so, because a percentage drawn from six items swings twenty points on one lucky guess.

If we ever publish a percentile estimate it will be because an independent validation study supports it, and the limitations will be printed alongside it. Our full position is on the methodology page.

How to use practice results properly

Track the trend, not the number. A single practice result is mostly noise — question selection, how much sleep you had, whether the topics happened to suit you. Four weekly results tell you far more, because the noise averages out while genuine improvement keeps showing up.

Expect different subtests to move at different speeds, and do not read a plateau as failure:

• Fast: Aviation Information and Word Knowledge, because both are recall.
• Steady: Table Reading and Block Counting, which improve with repetition over four to six weeks.
• Slow: Reading Comprehension, which tracks a habit built over years.

A flat trend across three weeks usually means you are practising rather than studying — taking test after test without going back to the material behind the misses.

The Self-Description Inventory does not score

Two hundred and forty statements, forty-five minutes, not graded, contributing to no composite. There are no correct answers and nothing to prepare. Answer consistently and honestly; instruments of this type are designed to detect people presenting an idealised version of themselves.

Why a subtest can matter twice, or not at all

Because composites are built from overlapping sets of subtests, the value of studying any one of them depends entirely on which composites you need — and the range is wider than candidates expect.

A subtest feeding several of the composites relevant to you is worth disproportionate effort, because every hour pays into multiple numbers a board will read. A subtest feeding none of them is worth nothing to you personally, however interesting or difficult it is, and time spent on it is time taken from something that counts.

This is why we keep returning to the same instruction: establish which composites your career field requires before you decide what to study. It is a single conversation with your recruiter, and it changes the plan more than any study technique will.

The corollary is uncomfortable but freeing. Most candidates can legitimately ignore several subtests entirely. Preparing for all twelve equally is not thoroughness; it is a failure to ask which ones count.

Why we will not give you a predicted score

This is the most common request we get, and the answer is the same every time: nobody outside the Air Force can produce one, and any site that hands you a number has manufactured it.

The reason is structural rather than a matter of effort. A composite percentile is produced by converting raw subtest counts through published-to-nobody conversion tables built on a reference population. Two pieces are missing to anyone outside: the conversion tables themselves, and the norming group your result would be placed against. Without both, a raw count of correct answers cannot become a percentile — it can only become a raw count of correct answers.

What a practice result can honestly tell you is comparative and internal: which subtests you are strongest and weakest in, whether you are running out of time and where, and whether this week is better than last week. All three are genuinely useful for deciding what to study, and none of them requires a percentile.

It is worth being clear about what the alternative would look like, because it is common. A site that shows you “estimated Pilot composite: 71” has picked a plausible-looking mapping from its own difficulty to a percentile scale. The number is internally consistent, responds to your performance, and means nothing — and a candidate who chooses whether to apply on the strength of it has been badly served.

Before you plan anything

Find out which composites your career field needs, then take the diagnostic and cross off every weak subtest that does not feed one of them. What remains, ordered weakest first, is your study plan — and the study guide covers how to sequence it against a test date.

For the structure of the exam itself, including what is scored and what is not, see the test format page.

What to do with a practice result

Read it in one order: pacing first, then accuracy by subtest, then the total. The total is the least useful of the three and the one people look at first — it aggregates away exactly the information that tells you what to do next.