For most of the Multiple Guess standardized Testing the correct question/answer pair is found by correlating that question/answer with others where the high sco

For most of the Multiple Guess standardized Testing the correct question/answer pair is found by correlating that question/answer with others where the high scoring test takers scored the same way... and differentiated from other low scoring test takers who get it wrong (some other answer). Testing houses have been using Big Data and correlation for many decades! As a result, if you take the questions as part of a meta test, you can make some assumptions about answers that are "too easy" and instead choose one which is more correct (because high scorers tend to use it). That often means that correctly used, obscure but field relevant, vocabulary is actually the right (highest scoring) answer choice.