CFA Practice Question

There are 676 practice questions for this topic.

CFA Practice Question

On a survey questionnaire, students were asked to indicate their class rank in college. If there were only four choices available, which measure of central tendency would be appropriate to use for the data generated by that questionnaire item?
A. mean and mode
B. mode and median
C. mean and median
Explanation: We can discard the mean, since it is not appropriate here. If we want to find out which class has the most students or where the students are most frequently ranked, the mode and median should be used.

User Contributed Comments 10

User Comment
mtcfa Why is the mean not important???
lilbut outliers..
bansal average rank is meaning less.
AusPhD Ignore lilbut - there are no outliers with such data. Bansal is correct
lanhuongnguyen class rank, miss that term...
Shaan23 Isnt the median rank useless as well. I'm thinking rank means Excellent, Good, avg and bad.

Mode would make sense here but not median. Class rank must mean something else I dont know.
Debashree i agree with shaan 23
NickGerli Perhaps if the class ranks were designated as percentiles...25th, 50th, 75th and 100th the median would be useful.
Benn09 So I get now why mean is bad, it is ordinal data and doesn't really give an interval if the mean isn't a natural number.

But although I can't eliminate them, I can't see why you would want them? Class rank is already curved to make the rank, so the median/mode should be incredibly predictable right?? And the number would only show your sample was a poor indicator, right? Doesn't seem useful
ascruggs92 ^Benn09 - don't think too hard man. Your making assumptions about a theoretical sample set here, no reason to do that. Mode would tell us which of the 4 groups most students fell into, and the median would tell us which group the middle observation falls into. Don't hypothesize about what class ranks in real life might look like, because that doesn't change the answer to this question.

NickGerli - the median would still be useless in that regard. Each group in your example wouldn't be a hard number, it would a range (i.e. 0-25, 26-50, 51-75, 76-100). How would you even calculate the mean at that point? Without knowing the exact value of each observation, the mean doesn't yield a meaningful value
You need to log in first to add your comment.