Friday, October 28, 2011

Multiple Choice Tests

The answer to this multiple choice question is
a) 42
b) All of the above
c) None of the above
d) None of the above

We just spent three days taking standardized tests. This being a public setting, I won't go into my feelings on the whole thing (this sentence probably says enough). I did spend some time before each test reminding my students that this was not a measure of how smart they were or how good they were at math, but only a measure of how many questions they could answer in a subset of a subject in a specific amount of time at a particular time of day on a certain day. This didn't make it any easier to see one of my thoughtful, deep thinkers frustrated after not finishing one section on time.

So much for not going into my feelings on the subject. This specific incident got me thinking...

Is there a psychometric reason for making tests timed and are tests intentionally constructed so that a certain percentage of students will not finish?

I did a cursory search of JSTOR and the interwebtoob and didn't find anything answering this question beyond what I see as a lame answer of "in the real world, you'll have tasks that will have to be completed with limited time".  True, but supposedly these tests are not testing your ability to complete tasks with limited time, they are testing things like mathematical reasoning.

Anyway, a question to the community.

Oh, and here's my second favorite multiple choice question* of all time courtesy of @ilovecharts

*My favorite multiple choice question isn't one particular question, but a self-referential test I saw years ago. A version can be found here.


  1. I love this question you've posted about multiple choice questions chosen at random.

  2. I love your 2nd favorite multiple choice question. I am a K-8 teacher candidate and am wondering how I will do with my students when it comes time for their standardized tests. I really value the question you ask about whether or not these tests are designed to leave some out.