Anyone annoyed by poorly written test questions in med school?

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

womp

Full Member
10+ Year Member
Joined
Apr 21, 2009
Messages
1,143
Reaction score
62
Our test questions are so poorly worded, and full of ambiguity that doesn't really test critical concepts. Just ambiguous for ambiguity sake. The answer distributions for certain questions are like 30% for one answer and 50% for the other choice. They don't require actual thinking so much as figuring out what the hell the question and the answer choices are even saying.

In comparison the BRS review questions are so much clearer and actually require some thinking.

The margin to honor an exam is just a couple of questions, so ugh.

People who go to pure pass/fail schools have it good, they just have to know their material for themselves and not worry about acing every poorly thought out question.

Members don't see this ad.
 
You don't like ambiguous questions?

Weird.
 
Members don't see this ad :)
I love the questions where it ends up being 20% for the correct answer, and about 60-70% for one other answer, but we're all still wrong and stupid for not getting the question right.
 
And there are still people who get 100% in spite of all this.

I don't think I could bubble 100% accuracy with the answer sheet in front of me. :scared:
 
The other peeve is the tight bell curve of the scores. The means are so high that 2 percentage points take you from the 90th percentile to the 75th percentile.

This makes poorly worded questions even more frustrating.

Wouldn't it make more sense to create a test that distributed people out a little bit better?
 
Just refer them to NBME exam question writing guidelines. And complain to the course director. That'll learn 'em.
 
Dealing with the same issue. I spent more than half time trying to understand the question on my exams.
 
Medical students should not be confused for a gold standard of knowledge.

Ive dealt with poorly written test questions, and Ive also seen plenty of whining about valid questions just because some people and their friends missed it. To make an argument that if x% of people put a particular answer, it becomes the correct answer is asinine.
 
The other peeve is the tight bell curve of the scores. The means are so high that 2 percentage points take you from the 90th percentile to the 75th percentile.

This makes poorly worded questions even more frustrating.

Wouldn't it make more sense to create a test that distributed people out a little bit better?

I thought we left the curving behind in undergrad. My school is ABCfail but not curved at all
 
I thought we left the curving behind in undergrad. My school is ABCfail but not curved at all
It may be and you may not realize it. I learned about the "stealth curve" in many of the courses at my school. Sure, they make a big deal that the grade cutoffs are absolute and that theoretically everybody can get an A, but they keep statistics on each individual question in their question bank (% answering correctly in previous years, etc) and then create tests with a set of questions that make it statistically likely that the results will approximate a bell curve with the desired mean. Couple that with the fact that many of these questions are ambiguous/poorly written (and often selected because they have the proper % answering correctly so that the average comes out right) and you get a bunch of annoying f***ing tests.
 
Sums up the first 2 years of medical school quite well, I must say.
 
Our test questions are so poorly worded, and full of ambiguity that doesn't really test critical concepts. Just ambiguous for ambiguity sake. The answer distributions for certain questions are like 30% for one answer and 50% for the other choice. They don't require actual thinking so much as figuring out what the hell the question and the answer choices are even saying.

In comparison the BRS review questions are so much clearer and actually require some thinking.

The margin to honor an exam is just a couple of questions, so ugh.

People who go to pure pass/fail schools have it good, they just have to know their material for themselves and not worry about acing every poorly thought out question.

The answer is "yes." Haha... To the extreme arguing degree!! Annoying if it takes class time, but very funny to watch occur in the hallway!
 
It may be and you may not realize it. I learned about the "stealth curve" in many of the courses at my school. Sure, they make a big deal that the grade cutoffs are absolute and that theoretically everybody can get an A, but they keep statistics on each individual question in their question bank (% answering correctly in previous years, etc) and then create tests with a set of questions that make it statistically likely that the results will approximate a bell curve with the desired mean. Couple that with the fact that many of these questions are ambiguous/poorly written (and often selected because they have the proper % answering correctly so that the average comes out right) and you get a bunch of annoying f***ing tests.

I could see that but a stealth curve is not a true curve and wouldn't cause the situation I was responding to (a few questions drastically changing your percentile), it would do the opposite by having a wider distribution. Kinda like Step 1
 
Remove yourself from the constant worry about grades. Your patients will only care if you can treat them effectively, so focus on becoming a good clinician--not test taker.
 
Questions with ambiguous answers that show skewed percentages are ridiculous. I had the best professor who after each exam would basically do a statistical analysis of each question and the results of each answer. Then he would build several charts, calculate several distribution equations and identify if incorrect questions were good distractors but not ambiguous. Then he would decide whether to throw out that question or use it again, but change the answer in some way to make it less ambiguous. Absolutely genius, if only other profs could act with that kind of professionalism.
 
You're ignoring the fun part is the fact that the unwritten rules varies by who wrote the question. One professor will act like anything not mentioned doesn't exist. Another will act like any conditions not mentioned may or may not exist. With a matching section, one professor will write questions where an answer choice can be used multiple times and another one will act like you can use process of elimination. This, of course, donesn't even touch on the issues of variations in different professors' lectures where the "right" answer varies depending who wrote it.
 
Questions with ambiguous answers that show skewed percentages are ridiculous. I had the best professor who after each exam would basically do a statistical analysis of each question and the results of each answer. Then he would build several charts, calculate several distribution equations and identify if incorrect questions were good distractors but not ambiguous. Then he would decide whether to throw out that question or use it again, but change the answer in some way to make it less ambiguous. Absolutely genius, if only other profs could act with that kind of professionalism.

The only problem with this is the use of med students as a gold standard for knowledge. I guess the bright side of this method though is abnegation from the responsibility of actually making a decision.:idea:
 
Kind of makes me feel better that there are a lot of us in the same boat.
 
Top