• Your scaled score is NOT a percentage correct…it’s like the SATs where the score is way more complicated than a simple division by 800 and is instead based on means and standard deviations normalized for how difficult your particular exam division form happens to be….a score of 518 means that you missed by two-ish questions.

• I’m trying to understand NCARB scoring as well. I recently took their practice exam and I decided to submit my exam unfinished, answering one question above their cut score. For PDD, the cut score ranges between 58-65 answered correctly, I got 66 out of 100 answered correctly. The result was a fail. It to my understanding these exams are scored on an overall items answered correctly. I’m looking for answers, if anyone have an understanding, I will very much appreciate an explanation. Thank you

• All of the questions have different difficulty levels. NCARB tests questions on test takers, and depending on the average performan. Theyey assign a difficulty level to each question. So all questions carry different specific weights, but still, every correct question earns you 1 point.

So if every question earns you 1 point, what does the specific weight do?

When you take an exam, the system randomly selects questions from a larger pool and represents them to you. Depending on the difficulty level of the question you get on a test combined, it calculates the number of the correct questions you need to answer to pass that specific batch of questions. The unique difficulty levels of questions are based on the performance of other test takers who answered the question before. Let's say 75% of the people got a question correct and you got that question on your test. And let's assume that most of the questions you got on your exam were answered correctly by most of the previous test takers. So this means you need to score higher on that test to pass because you got an "easier batch". If the passing score ranges between 58-65 you need to stay closer to the 65% side. Another critical factor is unscored test items. On top of getting a relatively easy batch, let's also say that you answered 66 of the 100 questions correctly but 8 of that 66 questions were test items and didn't count. This means that out of 91 questions that were scored, the correct ones that will be scored for you are now 58 questions. Which brought your percentage down to 63% and made you fail by 1 question.

But in the end, there is no way of exactly knowing how you performed on a given test because the scaled score only provides a general idea. The specific weights of each question can't be known for sure for the batch you received. But I still think a scaled score is more helpful than earlier versions of the score report when it comes to understanding your ballpark performance on a test. I actually created a calculator based on the scaled scores (https://arequestions.com/how-do-i-read-score-reports/) but as I said it is only an approximate result, no one can know it exactly with given data.

Hope this makes sense.