There are 165 questions, 15 do not count. Unfortunately, you don't know which ones do not count during the test.
Scaling the test allows for multiple versions and random generation of questions. This way, if another test taker is working beside you, they may not only have a different question order, their test may include questions not included on yours. Questions are not equal in complexity, therefore, results are likely weighted depending on the particular composition for your test.
Traditionally, 70% passing rate (i.e. 105 questions correct out of 150) has been a reliable benchmark for licensing tests. Apparently, in an effort to toughen standards, this has been raised to around 75% (112.5 correct out of 150). Scaling accomodates the probability of taking a test with a slightly higher question complexity compared to your neighbor, increasing the likelihood that you would not need to get quite as many questions correct, as long as you answered the harder questions satisfactorily. This is why a total number count of questions correct may vary depending on your luck at the 'random' test question set.
Scaling the test allows for multiple versions and random generation of questions. This way, if another test taker is working beside you, they may not only have a different question order, their test may include questions not included on yours. Questions are not equal in complexity, therefore, results are likely weighted depending on the particular composition for your test.
Traditionally, 70% passing rate (i.e. 105 questions correct out of 150) has been a reliable benchmark for licensing tests. Apparently, in an effort to toughen standards, this has been raised to around 75% (112.5 correct out of 150). Scaling accomodates the probability of taking a test with a slightly higher question complexity compared to your neighbor, increasing the likelihood that you would not need to get quite as many questions correct, as long as you answered the harder questions satisfactorily. This is why a total number count of questions correct may vary depending on your luck at the 'random' test question set.
