1.2 Practice Scores, Percentiles and Selection

Key Takeaways

  • A count correct and a percentage correct describe how you answered a particular practice set.

  • A percentile describes relative standing in a specified comparison group and cannot be calculated from a practice percentage alone.

  • Historical Defence sources describe aptitude screening; their description does not establish a current CFAT cutoff for a particular occupation.

  • If a CFAT result or retest matters to a serving member’s file, ask the applicable personnel selection contact what standard applies.

Last updated: November 2026

What a practice score can tell you

In the Canadian Forces Aptitude Test (CFAT) practice booklet, the answer key allows you to check which choices you got right. Start by recording your number correct. Add a denominator so the result makes sense when you return to it later. “Nine correct” could be almost the whole set or less than half of it.

On a 12-item drill, 9 correct gives 9 divided by 12, or 75% correct. A different drill with 15 correct out of 20 also gives 75%. Equal percentages do not establish that the questions were equally difficult. They also do not say whether you can complete a full test, since the drills may have sampled different topics or taken different amounts of time.

FigureWhat it meansWhat you need in order to use it
9 correctThe number answered correctlyThe drill and item count
75% correctThe fraction of the entire practice set answered correctly, expressed as a percentageThe correct count and total number of questions
75th percentileRelative position in a reference distributionA documented comparison group, comparable results, and the rule for ties
Selection thresholdA standard applied in a defined selection processThe official standard and instructions for that particular process

Why percentages and percentiles are different

Suppose your instructor runs the same exercise with a fictional group of 40 learners. You are looking at the score of a learner who got 9 of 12 correct. Of the 40 people, 24 scored strictly below that learner's score, four scored exactly the same, and 12 scored above it. If you calculate the share strictly below, you get 24 out of 40, or 60%. If you include everyone with the same score, you get 28 out of 40, or 70%. Both calculations describe this invented group under different definitions. In neither case do you rank the learner at 75% simply because nine correct means 75% accuracy.

The same score can sit in a different place in a different group: 9 of 12 can be near the middle for one class and near the top for another. Comparison only works if the groups took a comparable exercise and you know how the statistic was calculated. A practice app might display a percentile among its own users, but that number cannot be converted into a Canadian Armed Forces (CAF) rank without evidence that the score and comparison group apply.

Separate a skill diagnosis from selection

The Defence research description of the historical test mentions a minimum cut-off in the selection then in use. The CFAT Practice Version explains that scores could inform occupational options at that time. Those publications establish the historical role. They do not provide rules for deciding a current transfer or for converting a third-party quiz into an official result. The personnel selection directive describes the broader use of research-supported selection measures and refers to internal direction for their application.

For example, suppose a serving member is directed to write the CFAT after unsuccessful occupation-specific training. They do well on word problems but miss most spatial net items during practice. That is evidence to spend more practice time on rotating a marked face and checking edge connections. It is not evidence that the member does or does not meet any trade threshold. The applicable staff can explain what an assessment result means for their particular options. General online estimates, an old recruiting story, or a chart claiming a fixed percentile for a named trade cannot settle that question.

Use results to choose the next drill

  1. Count by skill and by cause. Record incorrect net items separately from arithmetic slips, vocabulary mistakes, and unanswered items. If you missed three spatial items because you reversed the visible side, that repeated mistake is a clearer starting point than your total score.
  2. Retest the method on fresh items. Redoing the exact question primarily shows whether you remember its solution. After correcting the face-orientation error, use an unfamiliar marked net to see if you can transfer the method.
  3. Compare on a fair basis. Keep question mix and conditions reasonably similar. A rise from 6 to 8 correct on comparable 10-item sets is an improvement of 20 percentage points on those sets. It is not a measured change in official percentile.
  4. Ask about any official retest directly. The Naval Experience Program instruction assigns certain potential retests to a Personnel Selection Officer. It does not establish a universal waiting period, eligibility rule, or which score governs a different file. Ask about authorization, preparation requirements and use of earlier results if a retest is offered in your case.

Official sources checked November 16, 2026.

Test Your Knowledge

A learner gets 9 of 12 practice items right. What can the learner determine without any comparison data?

A

Their official percentile is 75

B

Their official transfer options

C

Their percentage correct is 75%

D

Their score is above 75% of all candidates

Test Your Knowledge

In a 40-person practice group, 24 learners scored below a given mark and 4 more scored exactly at it. What share scored strictly below it?

A

24%

B

60%

C

70%

D

90%

Test Your Knowledge

A serving member sees an online chart with a CFAT percentile for a named trade. What is the appropriate way to check what applies to their file?

A

Use the percentile as the cutoff for every occupation

B

Translate the percentile directly to a percentage correct

C

Assume their best practice score will qualify them

D

Ask the applicable personnel selection contact which assessment and standard apply

Sections you finish are checked off in the contents.