I love MCQ exercises and exams. They are just so amusing sometimes. Take this one for example.
Fixed income securities promise...
A. a fixed stream of income
B. a stream of income that is determined according to a specific formula
C. either A or C is true
D. neither A nor B is true
Now assuming the question doesn't make sense to you, and assuming you're logical. The first assumption is just so that you ignore what the question tried to say and focus on what it did. The second assumption is so that you understand this question from a logical Boolean perspective. There are a couple of other assumptions but I think the most important is that there can only be one correct answer.
Therefore, since A or B is correct would imply that C is correct and vice versa, there would be direct contradiction of the assumption that there can only be one correct answer. So by computational logic, the "correct" answer is D.
Sadly it is not so.
But there are just so many loopholes in MCQ that it is kinda amusing sometimes. Instructions posed at the beginning of my paper go something like, "
Please write your answers on the answer sheet at the end!" To the untrained student, this means just writing the correct answers at the end of the paper. BUT to an experience student like me, I have to take it such that the teacher doesn't want the correct answers, but rather the teacher wants MY answers. Clearly here the teacher has intentionally asked for my answers, whether correct or wrong, and hence I will not be held liable for any flunking out of college due to miscommunications between teacher and student.
However, MCQ exercises also hold tremendous statistical power. I recall my first actuarial exam was MCQ, and I met a guy there who shared a little bit of trivia, that in 40 questions, there are like 7 A's, 9 B's, 13 C's and 11 D's (numbers not accurate due to fuzzing of memory) on average in the past year papers. So technically, since this is the average, therefore, when in doubt in such an exam, put C as an answer, and you'll have a statistically higher chance of getting it correct. MAYBE, since this was a statistics and probability exam, this was the
correct way for it to be done, rather than studying through textbooks on probability. After all, this would be a test on our creativity and statistical skills in collecting data for future application, and it would then be the perfect test. However, we must remember that the people administering these exams are geniuses at probability. Hence, we must observe that after an infinite number of papers, the distribution of correct answers should be approximately equal. So in future exams, there could be more A's and B's in the papers as compared to C's and D's. So either way, if you're taking the exam, it's your call.
But my statistical education has long been lost in the halls of the casino in Malaysia. And MCQ is so alike betting red or black on the roulette wheel, and in the off-chance both answers are wrong, the house wins. But MCQ still remains a pathetic way to test students, because in the off-chance that they decently picked up question-answering skills, they'd know... the longest answer is usually right, the most complicated answer is true and there is hardly ever any situation where you get 4 or more of the same answers in a row, and it is impossible to score less than 33% on an MCQ test, because anyone can at least omit one incorrect answer leaving them to choose one from three. But yeah, what do I know about statistics and MCQ?
You see, the biggest reason MCQs should not be used is because think about which possible realistic scenario can be construed into an MCQ. None! Imagine this:
Your wife asks you, "Does this make me look fat?" You respond in the following way,
A. "Yes" and sleep on the couch tonight
B. "No" and sleep in the garage tonight
C. Run out the door to the nearest bar to get drunk
D. Kiss her and remove the offending garment to have sex before you throw up.
See? MCQs are hardly valid. Just like Jeopardy questions, "Who is Darth Vader?" and they are suppose to respond with a question. Of course I could answer, "Not my father." but that's besides the point! There are so many answers to life, the universe and education that I'm somewhat surprised that it is possible to reduce such choices to four letters of the alphabet. But MCQs are here to stay, because they were made for the offendingly lazy... afte all... when there's only one right choice among four... what could be more simpler? Oh wait, true/false questions.