Using Rational Statistics to Improve “Educated Guessing”

So, you’ve spent the hours reviewing all of the important details for your board exam. You’ve done your best to target the weak areas, spending extra time focusing on topics that you didn’t quite master the first time through. And, perhaps most importantly, you’ve spent a considerable amount of time and effort doing practice questions and exams—all written by different authors—completed with time constraints in a variety of environments. By all measures, you are ready for this exam.

Once in the exam, you find yourself working a good pace, keeping up with all questions, and on target to finish with more than enough time to check your answers. And then, you find yourself in a position that we have all been in before…facing a question that is unfamiliar.

Despite all your preparation for this exam, you will undoubtedly find yourself facing a question that requires you to make an educated guess. This is an area of test taking that very few of us truly prepare for. Most often in this situation, we blindly pick an answer that “just feels right” or follow a hunch. And often times, when we do follow these hunches, we are wrong.

Successful question writers are known to put together challenging questions that less than 30% of test takers will get correct. (This is how exams develop that nice, bell-shaped curve of distribution that helps better clarify passing scores.) For these types of challenging questions, one or two answer choices may use a variety of techniques that appear correct and distract your attention. These techniques may include directly quoting a segment from the question stem or forcing the examinee to make a leap in logic to connect the answer to the question.

Knowing this, what is the best way to go about questions that require educated guessing? First, do everything possible to eliminate answers you know are incorrect. If you are not absolutely sure you can eliminate an answer choice, then it should not be eliminated. In these scenarios you want to be completely rational and play any statistical advantage in your favor. That is, if you have four choices and cannot eliminate any answer, you have a 25% chance of randomly guessing the correct answer. If you can eliminate one incorrect answer, you have increased your likelihood to 33%. However, if you eliminate the correct answer because you follow a hunch (as opposed being completely sure that it is incorrect), you have now reduced your likelihood to 0%. It doesn’t take a genius to know that 25% is greater than 0%.

After you have eliminated any and all known incorrect answer choices, the decision still remains: which answer to choose? Long before taking your exam you should determine your strategy. Will you pick the first of the remaining answers or the last of the remaining answers? Perhaps your strategy is to pick B, or, if it is not available, pick the next available answer. Whatever methodology you choose, decide before the exam and stick to that strategy once the exam begins.

Maintaining a rational thought process during the high stress environment of an exam is certainly a challenge that can impact your level of success. We can increase our chances for success in these situations by deciding what our educated guessing strategy is going to be and using statistics to our advantage by only eliminating answers that we are absolutely positive are incorrect. Following a hunch rarely works in games of risk—neither does it work during an exam.

Example MCQs and how to employ the technique:

Malarial parasites are treated with drugs containing:

1. Paludrine
2. Penicillin
3. Fentanyl
4. Aspirin

Answer: A

In this case, I can immediately eliminate answers C and D. I have now increased my likelihood of selecting the correct answer to 50%. If I could not distinguish between answers A and B then I would default to my personal pre-determined selection criteria, which is the first of the remaining answers. In this case, that choice is A, which happens to be correct.

Which of the following features of early multiple sclerosis are not predictive of long-term disability?

1. Frequent relapses during the first 2 to 5 years
2. Incomplete recovery from relapses despite therapy
3. Large lesion volume
4. Visual changes/Optic neuritis occurring within first 2 years of presentation
5. Short interval between relapses during the first 2 years

Answer: D

In this case, I have a 20% chance of randomly guessing the correct answer. Our attention is immediately drawn to answers A and E because they both contain familiar phrase. Again, selecting either of these choices based on hunch would have been the wrong decision.

Personally, I cannot eliminate any answers as I am largely unfamiliar with prognostic criteria for MS (leaving my percentage at 20%). Following my random guessing strategy, I would have selected answer A, which would have been incorrect. However, my choice was made on a pre-defined strategy, as opposed to hunch. Over the course of the exam, strategy is a far better technique to follow.

---

All of your Board test prep needs can be met by The Pass Machine!