How the CRPO Registration Exam Is Actually Scored (And Why It Changes How You Should Answer)

There is a strategy question every candidate eventually asks about the CRPO registration exam, usually late at night: "in a SELECT AS MANY section, how many am I supposed to pick?"

The honest answer is that the question itself is a misunderstanding, and the scoring system is why. Once you see how the exam is marked, the right answering strategy stops being a guess and becomes almost mechanical. This guide explains that system, drawing on the official Resource Manual for the examination (COMPASS, 2025), and finishes with what it means for how you should actually behave on exam day.

If you have not yet read our walkthrough of the exam's overall format, start there; this article assumes you know the basic structure of simulations, sections, and the two section types.

Once you understand the scoring, the next step is practicing against it. RegExam Prep calibrates all 615 of its options on the same +3 to -3 philosophy and shows you exactly what over-selecting costs you, section by section.

Every option carries its own score

Here is the core fact from which everything else follows: in every section of every simulation, each answer option has been independently assigned a value on a seven-point scale before you ever sit the exam (COMPASS, 2025). At the top of the scale sit options considered essential, where failing to select them would seriously compromise client care. Below those sit options that are strongly helpful, then mildly helpful, then neutral options that neither help nor harm. Below neutral, the scale turns against you: mildly detrimental, seriously detrimental, and at the bottom, options considered gravely damaging to the client.

These values are set in advance through structured review by panels of practising clinicians, and the passing standard itself is established through a criterion-referenced standard-setting process rather than by grading on a curve (COMPASS, 2025). You are not competing against the other candidates in your sitting. You are being measured against a fixed definition of safe, entry-level practice.

Two implications land immediately. First, there is no predetermined number of correct selections in a SELECT AS MANY section, because sections vary in how many positive and negative options they contain. Second, selecting a harmful option is not the same as wasting a click. It actively subtracts from your score.

The arithmetic of over-selecting and under-selecting

Consider a hypothetical Decision Making section, not from any real exam, with five options calibrated as follows: option A is essential (+3), option B is strongly helpful (+2), option C is neutral (0), option D is seriously detrimental (-2), and option E is gravely damaging (-3).

A candidate with sound judgment selects A and B and scores 5, the section's maximum.

An anxious over-selector, reasoning that "more coverage is safer," selects A, B, C, and D. Their score is 3 + 2 + 0 - 2 = 3. The essential and helpful actions they correctly identified have been partly cancelled by the harmful one they added. A candidate who selects everything scores 3 + 2 + 0 - 2 - 3 = 0, a complete wipeout of a section where they had, in fact, recognized every good option.

A timid under-selector, choosing only what feels absolutely certain, selects A alone and scores 3. Nothing was subtracted, but two available marks were left unclaimed, and across ten simulations those abandoned marks compound.

This is why the exam punishes both of the strategies that anxiety naturally produces. Clicking generously to be safe is not safe. Clicking minimally to be safe is not safe either. The only strategy the scoring system rewards is evaluating each option on its own merits: would this specific action serve this specific client at this specific moment? If yes, select it. If no, or if you cannot say why, leave it.

Section pass levels, and the totals that actually decide

Every section carries a minimum pass level, a benchmark score for that section established during standard setting (COMPASS, 2025). It is tempting to conclude that you must clear every section's bar individually, and mercifully, that is not how the exam works.

Your result is decided by two exam-wide totals: your total raw score across all Information Gathering sections, and your total raw score across all Decision Making sections. Each total is compared against its own aggregate benchmark. Falling short in one section can be compensated by strength in another, so a single rough simulation does not sink you (COMPASS, 2025).

But the compensation has a hard boundary, and it is the single most important structural fact about this exam: the two totals are independent gates, and you must clear both. A brilliant Information Gathering performance cannot rescue a failing Decision Making total, and vice versa. The exam is asking two distinct questions about you, whether you know what to find out, and whether you know what to do, and it demands a passing answer to each.

For preparation, this means tracking your practice performance in both categories separately. A candidate whose practice consistently shows strong inquiry but weak decision-making, or the reverse, knows exactly which gate threatens them, and generic "more practice" is a far worse plan than targeted work on the lagging category.

The simulation that does not count

One of the ten simulations on your exam is an unscored pilot, included to gather performance data for future examinations, and candidates are not told which one it is (COMPASS, 2025). This matters for scoring psychology more than for scoring math. If one simulation feels bizarre, unusually hard, or unlike anything you prepared for, it may well be the pilot, and the correct response is to do your best and move on without letting it contaminate the nine that count. Many a candidate has spiralled over a simulation that was never going to affect their result.

Once you understand the scoring, the next step is practicing against it. RegExam Prep calibrates all 615 of its options on the same +3 to -3 philosophy and shows you exactly what over-selecting costs you, section by section.

What your results will and will not tell you

Results are released several weeks after the sitting closes, historically in the range of six to eight. What arrives is comparatively spare: your outcome, framed against the two category benchmarks. What does not arrive is the diagnostic detail candidates naturally crave, which options cost you, which simulations went well, where the marks bled out. For successful candidates this hardly matters. For unsuccessful candidates preparing to rewrite, it matters enormously, because it means the diagnostic work has to happen during preparation, while feedback is still available, rather than after results, when it no longer is. We have written a full companion guide for candidates facing a rewrite.

What all of this means for exam-day strategy

The scoring system, once understood, dictates a short list of behavioural rules.

First, never select to reach a quota. There is no quota. The question "have I picked enough?" has no answer because it is the wrong question; the right one is "does this option, by itself, serve the client? ‍

Second, respect the negative options. They are not filler. They are calibrated harms, placed there because real clinicians sometimes do those things, and the exam is checking whether you would. When an option gives you pause, that pause is data.

Third, protect both gates. During the exam you cannot see your running totals, so the protection happens in preparation: practise both section types deliberately, and know before exam day which one is your weaker category.

Fourth, spend your caution before the click, not after. Selections are permanent, so the deliberation the notepad exists for happens first. Once you have clicked, take the information the exam gives you and keep moving.

Fifth, let the pilot be the pilot. One strange simulation is not evidence you are failing.

Once you understand the scoring, the next step is practicing against it. RegExam Prep calibrates all 615 of its options on the same +3 to -3 philosophy and shows you exactly what over-selecting costs you, section by section.

A note on practicing this properly

Scoring-aware preparation requires practice that scores the way the exam scores: options calibrated across the full seven-point range, section benchmarks, separate Information Gathering and Decision Making totals, and feedback that shows you precisely what over-selection and under-selection cost you, numbers the real exam never shows anyone. That is exactly what we built RegExam Prep to do, with twenty original simulations and full timed mock exams, and it opens for the fall cohort this September. The free foundation remains non-negotiable either way: the Resource Manual (COMPASS, 2025), the Professional Practice Standards (College of Registered Psychotherapists of Ontario, 2024), and the official practice exam in the weeks before your sitting.

The exam's scoring system is, in the end, a compliment to the profession. It assumes that clinical judgment can be measured one decision at a time, that harm matters more than effort, and that knowing what to ask and knowing what to do are different competencies deserving separate proof. Prepare in those terms, and the late-night question about how many to pick answers itself: exactly as many as the client in front of you needs.

References

College of Registered Psychotherapists of Ontario. (2024). Professional practice standards for registered psychotherapists. https://crpo.ca/wp-content/uploads/2024/09/Professional-Practice-Standards-Jan124.pdf

COMPASS. (2025). Resource manual: CRPO registration examination. https://www.compassexams.ca/wp-content/uploads/2025/02/Resource-Manual-CRPO-2025-_-ENGLISH.pdf