Many candidates finish a certification exam, see a number like 720 or 675 next to a passing line, and have no idea what that number actually means. It is not a percentage. You did not necessarily get 72 percent of the questions right. Understanding how modern certification exams are scored removes a lot of anxiety and helps you set the right preparation targets. This guide explains scaled scoring, why raw percentages are misleading, and how to use practice test results correctly.
Raw Scores Versus Scaled Scores
A raw score is the simple count of questions you answered correctly. If an exam has 65 scored questions and you got 50 right, your raw score is 50, or roughly 77 percent. The problem is that not every exam form is equally difficult. Vendors maintain large pools of questions and assemble different forms for different candidates. One form might happen to include harder questions than another.
To keep passing fair across all forms, vendors convert your raw score into a scaled score. Scaling adjusts for the difficulty of the specific set of questions you received. If your form was slightly harder, the conversion is more forgiving. If it was easier, the conversion is stricter. The passing bar on the scaled scale stays constant, so every candidate is held to the same standard of knowledge even though the raw number of correct answers needed can shift.
Why the Passing Number Is Not a Percentage
Consider a common example. An exam reports scores on a scale from 100 to 900 with a passing score of 700. That 700 does not mean 70 percent. It is a point on an arbitrary scale that the vendor defined during exam development. The relationship between raw correct answers and the final scaled number is set by a statistical process called equating, which most vendors do not publish in detail.
This is why you should never assume you need exactly 70 percent correct just because the passing scaled score looks like a familiar percentage. Depending on the vendor and the difficulty model, the actual raw percentage required to pass can be higher or lower than the scaled number suggests.
Unscored Pretest Questions
Most large certification exams include unscored questions, often called pretest or beta items. These are new questions the vendor is testing for future use. They are mixed in with the real scored questions and look identical, so you cannot tell which is which. They do not count toward your score at all.
If an exam is described as having 65 questions but only 50 are scored, the other 15 are pretest items. This is another reason raw percentage math is unreliable from the candidate seat. You do not know the exact number of scored questions, so you cannot compute your own percentage during the exam. The practical lesson is simple: treat every question as if it counts, because you cannot identify the free ones.
Comparison of Common Scoring Models
Different vendors use different scales and reporting styles. The table below summarizes patterns you will encounter. Always confirm current details on the official exam page, since vendors update these over time.
| Scoring element | Percentage style | Scaled score style |
|---|---|---|
| Passing bar | Fixed percent, for example 75 percent | Fixed scaled point, for example 700 of 900 |
| Adjusts for form difficulty | No | Yes, through equating |
| Result meaning | Direct fraction correct | Position on a defined scale |
| Predictable from practice tests | Somewhat | Harder to map exactly |
| Common in | Some foundational exams | Many associate and professional exams |
How Domains and Weighting Affect Your Result
Certification exams divide content into domains, and each domain carries a published weight. A domain worth 30 percent contributes more questions than one worth 10 percent. Some scoring models are compensatory, meaning strength in one domain can offset weakness in another as long as your overall scaled score clears the bar. Other exams enforce minimums per domain.
For most vendor certifications the model is compensatory across the whole exam, so you do not need to pass each domain separately. Still, ignoring a heavily weighted domain is risky. If a domain is worth a third of the exam and you barely studied it, you have surrendered a large block of potential points that the rest of your knowledge must make up.
Reading Your Score Report
After the exam you usually receive a score report that shows your total result and a breakdown by domain. This breakdown is often expressed as bands or general performance categories rather than exact numbers. If you fail, this section is the most valuable page you will get. It tells you which domains dragged you down so your next study cycle can target them.
Do not over interpret a single domain bar, though. With a small number of questions per domain, one or two unlucky items can shift a band. Look for consistent weakness across a full study cycle and multiple practice tests, not a single data point.
Using Practice Test Scores Correctly
Because scaled scoring is opaque, the safest strategy is to build a comfortable margin on quality practice tests. A common and defensible target is to consistently score meaningfully above the stated passing percentage on realistic, full length practice exams before you book the real thing. If the passing bar is described as 70 percent, aiming for steady scores in the low to mid 80s on good practice material gives you room for exam day nerves, harder forms, and the uncertainty of scaling.
Be honest about practice test quality. A practice test that is too easy inflates your confidence, while an unrealistically hard one can make you delay a ready attempt. Use official or well reviewed practice material, and always study the explanations for questions you miss rather than just recording the score.
Practical Takeaways
Scaled scoring exists to make certification fair across many exam forms, not to confuse you. Stop trying to compute your exact percentage during the exam, treat every question as scored, respect domain weightings, and read your score report as a study roadmap. Build a clear margin on realistic practice tests so that whatever scaling and difficulty your specific form brings, you are comfortably over the line. With that mindset the mysterious number on the screen becomes far less intimidating.
Frequently Asked Questions
Is a passing scaled score the same as a percentage?
No, a scaled score is a point on a defined scale, not a fraction of questions correct. A passing score of 700 out of 900 does not mean you answered 70 percent correctly. The vendor converts your raw correct count into the scaled number using a statistical process. This keeps passing fair across exam forms of different difficulty.
What are unscored pretest questions?
Pretest questions are new items the vendor is trialing for future exams. They are mixed in with real scored questions and look identical, so you cannot tell them apart. They do not count toward your result at all. Because of them, you cannot reliably calculate your own percentage during the exam, so treat every question as if it counts.
Do I need to pass every domain separately?
For most vendor certifications the scoring is compensatory, meaning your overall scaled score is what matters. Strength in one domain can offset weakness in another. Some exams do enforce per domain minimums, so check the official exam page. Even under compensatory scoring, ignoring a heavily weighted domain is risky because it surrenders many points.
What practice test score should I aim for before booking?
Aim to consistently score comfortably above the stated passing percentage on realistic full length practice exams. If the bar is around 70 percent, steady scores in the low to mid 80s give you a safe margin. This buffer accounts for exam nerves, harder forms, and scaling uncertainty. Always review explanations for missed questions rather than just tracking the number.