How the Next Generation NCLEX Measures Clinical Judgment

The exam is measuring your thinking, on purpose

Per NCSBN's 2026 test plans, the NCLEX is built on a clinical judgment measurement model that sits underneath the item bank. Not a question type. A framework for what the questions are trying to measure.

That framework is why the exam your program described a few years ago and the exam you will sit can look like different animals. The content did not shrink. The thing being scored moved.

Here is what the model is, where it sits in the exam, and which NCSBN document each claim below comes from.

What the NCJMM is, and what it is not

NCSBN calls it the NCLEX Clinical Judgment Measurement Model, usually shortened to NCJMM. It is layered. Each layer describes clinical decision making at a different level of zoom.

Layer 0 is the widest. NCSBN describes it as the broadest layer, encompassing all the clinical decisions a nurse makes to address a patient's needs.

Layer 3 is the one that touches your exam. NCSBN says it outlines the cognitive aspects of clinical decision making that are directly measurable and are the basis for the development of test items and case studies for the NCLEX exam. Layer 4 provides examples of contextual elements.

We are not going to describe Layers 1 and 2. NCSBN's current published material does not spell them out in a form worth quoting, and the plain-English versions in circulation trace back to third-party summaries rather than to NCSBN. Where the source is thin, we say so instead of filling the gap.

Layer 3 carries six cognitive operations. Recognize cues, analyze cues, prioritize hypotheses, generate solutions, take action, evaluate outcomes. Both 2026 test plans give each one a one-sentence definition, and the wording is identical for the RN and the PN exam.

All six, with NCSBN's own definitions, sit in the six steps explained without the jargon.

Why judgment replaced pure recall as the target

NCSBN's clinical judgment page describes the model as built from research involving over 100 nursing experts and analysis of data from more than 200,000 NCLEX candidates. That is the stated basis. Worth knowing, because the change gets described online as a marketing exercise, and it is not one.

The practical argument is easier to feel than the research one. Recall asks whether you can retrieve a fact. It does not ask whether you would notice the fact mattered.

A nurse who can recite every sign of low potassium and still walks past a monitor that has started throwing U waves has passed the recall test and failed the shift. The exam is trying to catch that gap before a license does.

Recall is still necessary. It is just no longer sufficient on its own, and the item formats exist to force the extra step.

How case studies and stand-alone items divide the exam

Both 2026 test plans state that every exam carries three clinical judgment case studies. Each case study is a set of six items. That is 18 items in total, mapped to the six operations in Layer 3, and the wording is the same for RN and PN.

Two details inside that sentence do real work.

The first is arithmetic. Those 18 case study items are counted independently of the Client Needs percentages. They sit on top of the content distribution rather than being carved out of it. The category weightings you plan your weeks around are not quietly competing with case study volume.

The second is that case studies are not the only home for judgment. NCSBN states that approximately 10 percent of a candidate's exam is made up of stand-alone clinical judgment items, selected depending on exam length. Those arrive looking like ordinary items.

So judgment is measured in two places at once. A concentrated block, and a scatter. If you have only practiced case studies, you have prepared for one of them.

How a case study is assembled, unfolding client record and all, is covered in the case study format explained.

The formats, and where the arithmetic lives

NCSBN's current item-types article names five format families: extended multiple response, extended drag and drop, cloze drop-down, enhanced hot spot highlighting, and matrix or grid. Bowtie is named separately, in NCSBN's scoring FAQ.

You will notice we are not walking through each one here. That is deliberate. Each format gets its own page, because each one fails differently under time pressure.

The five-point structure of a bowtie item is a different problem from the reading order for a drop-down cloze stem, which is different again from a drag and drop task that hands you more options than spaces.

Scoring is its own subject. Partial credit is real, and whether a wrong selection costs you anything depends on which model the item uses. That arithmetic belongs in plus-minus scoring and the cost of one extra selection.

One item type you may have seen listed elsewhere is missing from all of this on purpose. We name what appears in NCSBN's current documents and nothing else.

How to tell a judgment item from a recall item

Recall items ask what. Judgment items ask what now, and they hand you a situation instead of a term.

A recall item asks which finding goes with a condition. A judgment item gives you a client, a set of findings, and asks which one you would act on first, or which finding contradicts the working explanation. The content underneath can be identical.

That distinction changes what a wrong answer means. Missing a recall item tells you to reread a topic. Missing a judgment item usually tells you the reasoning broke somewhere in those six operations, and the topic was never the problem.

Where to start if the model still feels abstract

Start at the seam between the first two operations. Recognizing a cue and analyzing it are separate steps, and candidates who merge them lose points inside case studies without ever knowing why.

That split is worked through on a single chart in recognizing cues versus analyzing cues.

Exam length and the rules that end your exam are a different machine entirely. Nothing in the judgment model tells you when the computer will stop, and that question is answered in how computerized adaptive testing decides your NCLEX.

The plan these facts come from is dated on purpose

NCSBN revises the test plans on a three-year cycle. The current NCLEX-RN and NCLEX-PN plans are both effective April 1, 2026 through March 31, 2029, and both were built on the 2024 Practice Analysis plus a 2024 KSA survey for each exam.

Check the date on anything you read about this exam, including this page. A test plan is a dated document, and prep material written against an older one is describing an exam that has since been revised.

Which document says what

Every claim above is traceable, so here is the trail. The six operations, the layer descriptions, the three case studies of six items each, and the approximate 10 percent of stand-alone judgment items all come from those two 2026 test plans.

The format names come from NCSBN's current help-center article on item types. The bowtie maximum and the three partial-credit scoring models come from NCSBN's scoring FAQ. The research basis for the model comes from NCSBN's clinical judgment page.

None of that is insider knowledge. It is published, and you can go read it yourself.