A proposal for continuously scaled grades

Instead of the letters A through F, why not have grades represented by a decimal from 0 to 5 (e.g. 3.7). This way, subtle gradations could be expressed, while the general quality of the grade could still be seen from its integer part. Thus anything in the 4’s is similar to an A in the current system, in the 3’s to a B, etc.

Why round the grades off to discrete values and lose information?

Also, in the interests of combating grade inflation, I would go so far as to calculate normalized grades, which would essentially represent the number of standard deviations of the regular grade from the mean for a given class. So a normalized grade of 1.5 would be one and a half standard deviations above the mean, while -0.8 would be 0.8 standard deviations below. Thus in a class where the professor gave all high grades in the 4.5 to 5 range, the better students could still be distinguished based on the normalized grade.

Just some thoughts: what am I missing?

Advertisements

About marcpevans

I'm a composer, and a graduate student at UC Santa Barbara. This blog contains my philosophical musings on music and on other things. If you want to actually listen to my music, you can find it at www.marcevansmusic.com. Welcome! :)
This entry was posted in Uncategorized. Bookmark the permalink.

4 Responses to A proposal for continuously scaled grades

  1. Nick Norton says:

    I dig the idea – but you know in the UC system that pluses and minuses go into GPA, right? B+ is 3.3, B is 3, B- is 2.7.

  2. ddrkirbyisq says:

    As humans it is often much easier for us to understand and categorize things into “really good” vs “sort of ok” vs “not so good” vs “really bad” than it is for us to use a scale from 1 to 100. The reason that sites that yelp use a 5-star system is because it’s much easier for people to ask themselves “was that restaurant a 3-star restaurant? Or was it a 4-star restaurant?” than it is to think about whether they should give it a 50%, 60%, 55%, 53.21%…

    It’s a similar reasoning for grades, but your argument is still valid since a lot of grading is done based off of quantitative values. I don’t think you are gaining much though…when I look at your grades I don’t actually care about the fine gradations; I mostly just want to know what your most common/median grades are and whether you have any really bad ones. What you said is only really useful when you are attempting to do a comparison between two different students, and in that case I don’t know if it feels better to let a minor point differential impact the result instead of just saying they have about the same grades.

    Regarding grade “normalization”, isn’t that the objective of grading on a curve?

  3. AGE says:

    Marc, you are not “missing” anything.

    As a former university (assistant) professor (rather than music I was in Veterinary Medicine) I always wanted nothing more than for my students to succeed. I lived by the concept of reward a (somewhat relative) “A for effort”, yet that was not truly being fair to the students that not only put in the effort yet also succeeded in learning.

    I totally agree with your concept to normalize grades based on a professor’s tendency to give “all high grades” … which was mine.

    Still, wouldn’t it be lovely if we could encourage all to thrive and succeed without a system based on judgements – normalized, rationalized, agonized … whatever.

    & have you ever heard the Question: What is the difference between the med student that was 1st in his/her class and the med student that was last in his/her class. Answer: Nothing. They are both “Doctors”.

    Smiles
    AGE

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s