No norm group
A percentile only means something relative to a defined population. Our bank is answered by a few thousand visitors on a website, self-selected, unproctored, and taking untimed sets. That cannot serve as a reference group.
How ASVAB scores actually work
Raw score, standard score, AFQT percentile, line score, stanine - five different numbers that people routinely confuse, which is exactly why so many prep sites get away with printing invented scores.
The chain
1. Raw score
The count of items you answered correctly in a subtest. That is the only number on this page that a practice session can legitimately produce.
Nothing else about a raw score transfers: it depends on how many items were administered, which form they came from, and - in a computer-adapted test - which items the algorithm decided to serve you. Two people with the same raw score on the same screen are not necessarily at the same ability, which is the entire reason the next step exists.
2. Standard score
The raw score is converted onto a common scale so different forms and different subtests can be compared. That conversion uses the norming sample - the population the test was calibrated on - and the equating model that links a form to the scale. Standard scores are the inputs to everything that follows.
3. The AFQT percentile
Four standard scores - Arithmetic Reasoning, Mathematics Knowledge, Word Knowledge, Paragraph Comprehension - are combined and the result is expressed as a percentile against the reference group. It is a ranking. It is not a percentage of questions correct, it is not out of 100 in the sense a school test is, and the same raw performance can land differently across forms and formats.
4. Service component and line scores
The other six subtests do nothing for eligibility, but they combine into composites that the services use to decide which jobs you can be assigned to. These are the numbers that actually shape your options once you are eligible, and each branch names and builds them differently.
5. Stanines
Some reports show a stanine: a band from 1 to 9 produced by collapsing the standard-score distribution into coarse groups. Stanines are deliberately low-resolution, which makes them useful for describing a general level and useless for tracking a one-point change. If a report you are reading offers a stanine derived from a practice test, it has invented the intermediate steps.
What ASVAB Pilot reports, in full
- Correct answers and total answered, per session.
- Accuracy percentage - correct over answered, nothing else.
- Time spent, against the section clock.
- Per-subject and per-topic breakdowns, so you know what to drill next.
- The share of a mock you had already seen, because familiarity inflates results.
What we will never print
No predicted AFQT, no converted line score, no stanine, no eligibility statement, and no claim that a practice percentage means you qualify for a service or a job. Accuracy of 70 percent here is 70 percent of our questions. It is not an AFQT of 70, and a site that tells you otherwise is guessing.
Why the conversion is impossible
Three reasons, honestly
No equating
Official forms are statistically linked so a score on one form means the same as a score on another. Our items are not calibrated that way - difficulty labels are editorial judgements until real answer data says otherwise.
Not adaptive
The computer-administered ASVAB changes what it serves you based on your answers, so difficulty is part of the score. Our timed sections draw randomly from a fixed pool and add a clock. That is CAT-style practice, not CAT scoring.
Clock reference
Sections that feed each number
Useful when you are deciding where to study: the AFQT four in bold, everything else feeding composites.
| Subtest | In the AFQT | Items | Minutes |
|---|---|---|---|
| General Science (GS) | Composites only | 15 | 12 |
| Arithmetic Reasoning (AR) | Yes | 15 | 55 |
| Word Knowledge (WK) | Yes | 15 | 9 |
| Paragraph Comprehension (PC) | Yes | 10 | 27 |
| Mathematics Knowledge (MK) | Yes | 15 | 31 |
| Electronics Information (EI) | Composites only | 15 | 10 |
| Auto Information (AI) | Composites only | 10 | 7 |
| Shop Information (SI) | Composites only | 10 | 6 |
| Mechanical Comprehension (MC) | Composites only | 15 | 22 |
| Assembling Objects (AO) | Composites only | 15 | 18 |
Item counts and clocks are the ASVAB Pilot practice template, not official administration rules. The official sources are the only authority on what you will sit, how long each section runs, and what your scores mean for a specific job.
GSAFQT
General Science training plan
15 items in 12 minutes, with the skills in drill order and a week-by-week routine.
ARAFQT
Arithmetic Reasoning training plan
15 items in 55 minutes, with the skills in drill order and a week-by-week routine.
WKAFQT
Word Knowledge training plan
15 items in 9 minutes, with the skills in drill order and a week-by-week routine.
PCAFQT
Paragraph Comprehension training plan
10 items in 27 minutes, with the skills in drill order and a week-by-week routine.
If a score report you received somewhere else is unclear, ask your recruiter for the official worksheet and the composites for the jobs you want - that is the document that governs. For anything about ASVAB Pilot, mail [email protected].
Measure the thing you can actually control
You cannot practise toward a percentile. You can practise toward fewer topic-level misses - which is what our reports show. Original questions, real explanations, no invented scores.
Download on theApp StoreAndroidComing soon