How NCLEX CAT Works: How Items Are Chosen and When the Exam Stops

What NCLEX CAT estimates after every answer

Computerized adaptive testing, or CAT, does not give every candidate the same fixed form. Each time you answer an item, the computer re-estimates your ability from all your responses so far and the known difficulty of those items. With each answer, the estimate becomes more precise.

Three details shape how that works on the NCLEX-RN:

• Everyone starts low. NCSBN's FAQ says each candidate first receives an item of relatively low difficulty, and the path from there depends on performance.
• Some items don't count. Every exam includes 15 unscored pretest items. They look like the scored items, so you can't tell which ones they are, and they are left out of the ability estimate and the pass-fail decision.
• The target is a line, not a percentage. The result depends on where your final estimate sits relative to the passing standard, which NCSBN has set at 0.00 logits for the NCLEX-RN through March 31, 2029. There is no percentage-correct cutoff.

That is why "I got most of them right" can't be checked from memory: the count means little without the difficulty of each item. Read how the NCLEX-RN passing standard works for the ability scale itself.

The NCLEX-RN limits CAT works inside

The 2026 NCLEX-RN Test Plan and the Candidate Bulletin set the frame for every exam:

The plan is in effect from April 1, 2026 through March 31, 2029, so these limits cover every 2027 test date as well.

Three criteria choose each NCLEX item

The Candidate Bulletin and the test plan describe three criteria for picking the next item from the pool:

1. Content area. The item has to come from the area that best keeps your exam on the test-plan percentages, so every exam gets enough items from each area. Three case studies and about 10% stand-alone clinical-judgment items are selected as well, depending on exam length.
2. Information. The item is targeted at your current estimate, so you should have about a 50% chance of answering it correctly. Items that are far too easy or far too hard tell the computer little.
3. Exposure. Items a repeat candidate saw on an earlier attempt are excluded.

Two consequences follow. A run of items from one topic reflects content balancing and targeting, not a report on your weak areas. And an item that resembles an earlier one is not a sign you missed the first: the test plan says similar items can address different phases of the nursing process, or one of them may be an unscored pretest item.

Because every exam has to cover all eight Client Needs areas, practice all eight Client Needs areas one at a time or mixed → rather than betting on the topics you expect to see.

Why NCLEX items keep feeling hard

If the computer keeps targeting items you have about an even chance of answering correctly, most of the exam should feel challenging. NCSBN's FAQ says both high- and low-ability candidates find the items near the end hard, and that candidates can't reliably tell which items are easy or difficult on the NCLEX scale.

The FAQ also says there is no set percentage of item formats. Getting more select-all-that-apply items, or a hard-looking last item, is not a result signal.

Stopping rule 1: 95% confidence

After you reach the 85-item minimum, the exam can stop as soon as the computer is 95% certain that your ability is clearly above or clearly below the passing standard. NCSBN calls this the most common way an NCLEX ends.

In plain terms, the computer asks whether the remaining uncertainty around your estimate still overlaps the passing standard:

• if the whole range sits above the standard, the result is a pass.
• if the whole range sits below the standard, the result is a fail.
• if the range still crosses the standard, the exam continues while items and time remain.

The 95% is confidence in the decision. It is not 95% correct, a required score or a pass probability. The test plan states that exam length is not an indication of a pass or fail: an exam that stops at 85 items can end either way.

Stopping rule 2: maximum length

When an estimate stays very close to the passing standard, the exam can run to the 150-item maximum. At that point, the computer stops using the confidence rule and looks only at the final ability estimate from all scored items:

• at or above the passing standard is a pass.
• below the passing standard is a fail.

Reaching 150 items means the estimate was close to the line, not that it was below it.

Stopping rule 3: run out of time (R.O.O.T.)

The run-out-of-time rule applies only when the five-hour clock expires before 150 items and before a 95% decision:

• fewer than 85 items answered is a failing result.
• at least 85 items answered means the final ability estimate from all completed items decides the result, pass at or above the standard and fail below it.

Finishing all 150 items falls under the maximum-length rule, even if little time is left. The guide on what happens if you run out of time on the NCLEX covers the timer-specific question and pacing.

What CAT should change about your NCLEX prep

Don't practice by trying to decode whether a question is "above passing level." The Candidate Bulletin warns that rapid guessing can drastically lower your score on an adaptive test, because wrong answers bring easier items and push the estimate down. Its advice is to keep a reasonable pace and read each item carefully before answering.

A practice test can't copy NCSBN's adaptive item selection, but it can build the stamina the real session needs: rehearse five hours of focus with an 85-question timed mock exam →.

After any practice set, review each item by asking:

• Which cue changed the risk?
• Why is the chosen action safer than the alternatives?
• Which distractor depended on an unstated assumption?
• Which Client Needs area or clinical-judgment step needs another look?

Those questions turn a practice score into a study decision. To work through them item by item, read the explanation for every answer, including the ones you got right →.

Before test day, check the NCLEX exam-day rules for breaks, the on-screen calculator and how submitted answers work.