Friday, September 18, 2009

What is standards-based grading?

It's probably easiest to describe standards-based grading in comparison to traditional grading. Yesterday, as an introductory lab, I asked the kids to try to figure out how fast a toy car goes. My kids didn't have any real background knowledge yet, but I did provide them with a meter stick and stopwatch.

Traditional grading:
In traditional grading I might have made this lab worth 50 points. Usually in science we require a lab report and assign point values to each section of the lab report. For example, I might have made a correctly written hypothesis worth 10 points, the procedure worth 5 points, and graph and data table worth 15 points. In the end, a student would receive a grade based on their lab report and some sort of point amount for "getting the right answer" that s=d/t.

There are many problems with this grading system and I will comment on them as I progress in this blog. I think the main problem is the misalignment of the purpose and the grading of the lab. The purpose of the lab was for students to attempt to figure out how to calculate the speed of a toy car and hopefully create a general rule or formula. What did I grade the students on? I graded them on how well they could follow a pre-assigned lab format. Imagine a student who was able to both calculate the speed of a toy car and figure out that s=d/t but for some reason decided not to number his procedure. On the other hand, you could have a student create the most beautiful data table in the world but completely miss the point of the lab. It's this misalignment of purpose and grade that is the primary failing of traditional grading.

Standards-based grading:
Here's where I always lose the holdouts on the teaching staff when I present standards-based grading to them. In standards-based grading, you only grade how well they have mastered a specific standard or topic. The grade completely ignores the format of the lab report1. Their grade is aligned with the purpose of the lab. The grade should answer the question,"How well do they understand the concept and calculation of speed?"

The grade reflects their level of understanding on that specific topic or standard and nothing else. It seems obvious, but in practice it's a huge paradigm shift.

1. Well, you might actually have a separate standard for writing a lab report in which case you would give two scores, one for the content and one for the format.

1 comment:

  1. OK. So I started reading your stuff this week (October 2012). And I put you on my blogroll right away with the following description:
    Holy cow! This dude has put into writing the thoughtful versions of my private assessment rants. Go get your assessment house in order by reading him from beginning to end.
    Of course 90% of those who might read my stuff have been following you for a while now. Be that as it may, it is truly heartfelt and I'm taking my own advice. Reading from beginning to end. It'll take a few lunch hours, I can see.
    You write, "The grade completely ignores the format of the lab report". This is so, so important, right? The grade is based on the knowledge represented in the work they submit. I get around the whole issue by having a minimum standard for acceptable work. If the format of the work doesn't meet that standard, it needs to get revised before it can get graded. I don't grade the format, but if it's not in an acceptable format I'm not going to grade it at all. This "good knowledge, bad writing" doesn't score the same as "good writing, bad knowledge". Bad writing gets revised.
    And then, of course you go the extra step and make sure that bad knowledge can get revised too. I aspire to that, but I'm not there.