scoring

How Aptitude Test Scores Are Calculated (And Why Percentiles Beat Raw Scores)

Aptitude test scoring confuses even experienced HR staff. The pipeline from a candidate's correct answers to a go or no-go hiring decision has several stages, and understanding each one is the difference between interpreting your score correctly and misreading it. The short version: raw score is step one of four, percentile is the number that actually matters, and role-mapped target scores are what you need to clear. Here is the full breakdown.

By Junaid Khalid, updated 2026-04-18

Key takeaways

  • Raw score is step one of four in the scoring pipeline.
  • Percentile is the comparable number across tests and norm groups.
  • Role-mapped target scores are what you actually need to clear.
  • Norm groups can be general, role-specific, or custom to the employer.
  • Confidence intervals matter. A 28 plus or minus 3 is a range, not a point.

Step 1: Raw score

Raw score is the count of correct answers. On a 50-question test, it ranges from 0 to 50. On a 30-question test, from 0 to 30. Simple and easy to understand.

What raw score cannot tell you on its own is how good the score is. A 28 could be average or above average or exceptional depending on the test, the norm group, and the role. Raw score is the starting point, not the interpretation.

Step 2: Norm group comparison

The raw score is compared against a reference population called the norm group. The comparison produces a percentile, which expresses your rank relative to the population.

Norm groups are chosen by the vendor and sometimes customized by the employer. Common norm groups include general applicants, role-specific applicants (e.g. software engineering candidates), country-specific applicants, and education-filtered populations (e.g. university graduates only).

The norm group choice has enormous impact. A raw score of 28 on the CCAT might be the 68th percentile against the general applicant norm group and the 45th percentile against a norm group of consulting applicants. Always check which norm group the employer uses.

Step 3: Role mapping

Employers set role-specific cutoffs. Some use the vendor's default role bands such as the entry-level, skilled, managerial, technical, and executive tiers that Wonderlic and Criteria publish. Others set custom cutoffs based on their own top-performer benchmarks calibrated to the specific role.

The role-mapped target is what you need to clear. It is rarely disclosed to candidates directly, but vendor documentation provides reasonable estimates for role families. The Predictive Index Cognitive Assessment is unusual in that the Job Target Score is explicitly set by the employer within the test configuration.

Step 4: Decision

Either you pass the cutoff or you do not. Some employers treat the cutoff as a hard gate. Others use it as one signal among several, weighted alongside resume screen, interview performance, and references.

The weighting varies widely. At some employers, the cognitive score is 50 percent of the hiring decision. At others, it is 15 percent. Internally, the cutoff itself is often a hard floor: you must clear it to be considered at all, and above the floor, other factors matter.

Why raw score is misleading in isolation

A raw score of 28 on the CCAT is meaningless without knowing the average is 24 and the 90th percentile starts around 38. A raw score of 28 on the PI Cognitive Assessment is completely different because the scoring scale goes from 100 to 450.

This is why percentile dominates actual hiring decisions. Percentile collapses the raw score into a single interpretable number that survives comparison across tests. When anyone refers to their score casually as "I got a 28," ask what percentile that represents. That is the real signal.

Confidence intervals and true ability

Most vendor score reports include a confidence interval around the point estimate. A score of 28 plus or minus 3 means your true ability most likely falls somewhere between 25 and 31. The interval exists because any single test session is noisy, and the true score is a range rather than a single point.

This matters for interpretation. If your target cutoff is 27 and you scored 28 with a plus or minus 3 interval, you are within the range of true ability that could be below the cutoff. Employers who understand confidence intervals sometimes set cutoffs above the interval to account for this.

Percentile lookup tables

Vendor documentation publishes percentile-to-raw-score conversion tables for most tests. You can look up your raw score to see the corresponding percentile for the vendor's default norm group. Employer-specific norm groups are usually not published.

If you have access to your raw score and the employer uses a custom norm group, you can estimate the difference by assuming their norm group is usually stronger than the general applicant pool. Expect your percentile against a custom norm group to be lower than against the general pool.

FAQs

Think in percentiles. Plan around cutoffs.

Benchmark your raw score against a realistic test right now.

Start a Free Practice Test