14 December 2007

Number Crunching

There is a grail which still needs to be found in standards-based grading. Mind you, O'Connor and others suggest that instead of searching for this grail that we should eliminate the need to look for it. While it is my hope that will one day happen, teachers need an alternative in the meantime. What the heck am I talking about? I'm talking about using standards-based grading practices within a traditional reporting system.

The problem is simply that teachers who use standards-based practices are organizing information by target---their gradebooks have no need to number crunch. Traditional report cards, however, do. I might talk with students about their progress toward the various learning goals, but I still have to put an A, B, C, D, or F on the report card. I would love to have a report card where I could share with parents how kids were progressing toward the individual goals (and thus eliminate the need to hunt for the grail). Right now, it isn't a possibility.

At the Sound Grading Practices conference, there was a session on converting rubric scores to grades. I had high hopes for this session---because I thought it might be the grail. Alas, it was a mirage. The idea was to use "logic" to just assign grades and/or percentages to the rubric scale. (E.g., let a 4 = A/95%, 3 = B/80%...) I didn't like this for two reasons. One, it isn't logical at all. There are no rules applied, but rather the teacher is somewhat arbitrarily assigning symbols. If that's the case, then why bother with the rubric scale at all? If you're going to say a 4 = A, then just put A, B, C, D on the rubric. (Ditto for percentages.) Secondly, I've done a lot of work getting kids to understand that a "3" means that they have met the standard and that is a "good" grade. My kids are excited by 3's. They know that a "3" is not a "B" (or any other letter equivalent). Why would I confuse them now?

One method, however, might have some promise. Here is one version of it:

In other words, as you look at the median scores for content and the most recent info for skills and are trying to report this out as a single letter grade (because you have to), then maybe a table like this would help. Is it possible to write an Excel algorithm for it, I wonder?

Any other ideas about how to engage in the Art of the Impossible and number crunch with standards-based grades?


Clix said...

You're pinching my BRAIN.

The Science Goddess said...

Yeah, but it's a good kinda hurt. Feel the burn, baby.

Clix said...

So what would you do if you had 25% 3s, 50% 2s, and 25% 1s?

The Science Goddess said...

Flip a coin. Heads is a "D" and tails is an "F," natch.

Seriously---this is a work in progress. If a kid is only getting 25% 3's (at standard) and 75% is below that, a "D" would be the highest possible grade.

Hugh O'Donnell said...

SG, you are so right about trying to quantify rubrics. After the session with Judy Arter, I commented to her that it must hurt, having to reduce rubrics to numbers we can crunch.

Judy, ever the diplomat, said that it was difficult, and I got the impression she was looking to put out the least damaging alternative.

The fact is, she would rather not do the "logic" trick, or anything else, and I'll bet that it won't be long before you see ATI recommending the gestalt "professional judgment" method of integrating rubrics into report card grades.

You've addressed the gestalt notion before, and that's where I'm at with it.

I really believe you have nailed it (quantifying rubrics) with the descriptor, "Art of the Impossible." Rubrics were fashioned to eliminate number crunching, so it seems to me that trying to loop them back around to a letter grade or a crunched number is to fashion a nonsensical Moebius Strip, one on which we keep coming up with no conclusion, no answer, always returning to our unanswerable question.

And, regardless of how we quantify, we're still dealing with cut-off points, and those are always arbitrary, unless we choose to define them in policy. And then they're still arbitrary.

I do admire the way you have charged that windmill, though! :)

Hugh aka Repairman

The Science Goddess said...

That's me...Donna Quixote, at your service.

Until report cards catch up with grading, we have to find some tangible way to determine grades. Something we (parents, kids, teachers, colleges...) can live with.

Athena said...

I was just going to email you over the weekend asking this very question! How do I report their grades in our computerized gradebook? We have to have 2 grades a week and send progress reports at midterm. I could see the table being used at the end of the term.

But how do I put the "two grades a week" in?

(I know--I just complicated the issue. Sorry.)


The Science Goddess said...

Not to worry---this issue is already complicated. LOL

In terms of how I put in my weekly grades, instead of categories like "tests, labs, homework...," I have "1.3.10: Ecology, 2.1.2: Investigative Design, etc." The grades can be put in the same way as a traditional set-up---it's just that the categories are different.

If the gradebook is just for you to track scores (and who has turned in what), then this should work. At the end of term (or progress report time), just look at the columns you need, rather than the automatic average calculated at the end.

The alternative suggested at the conference was to create a scale for A - F (and as many + and - as you want) that corresponds to the average score. I'm not personally comfortable with this as it treats standards-based scores as if they were traditional grades, but I can see that it might be an intermediary step for many teachers until something better comes along.

Navymom said...


I have recently become a standards-based grading convert. I have done as much reading as I possible can on the topic ( Marzano and Stigins). However, I have yet to find a source that actual helps in the implementation into a working classroom. Do you have any direction that you can give me. I am focusing on getting the gradebook organized, getting the admin/ parents/students on board, and the communication of the grades. Any help would be greatly appreciated

The Science Goddess said...


There is no "implementation guide" out there, as such. For more background on best practices in grading, I also recommend "How to Grade for Learning" by Ken O'Connor and "Grading" by Susan Brookhart. There is also an ASCD Inquiry Kit on Grading which has some good resources for a literature circle.

Karen Kay Harvey is a consultant who has worked with several districts on getting standards-based grading and reporting implemented. She was good enough to share her thoughts at the Sound Grading Practices conference earlier this month about how to best shepherd the beast. As with anything affecting education, change is slow...and not always welcome. There are other teachers, schools, and districts who are also in the throes of making changes in grading practices that you could make contact with. (E-mail me for more at the_science_goddess[at]yahoo[dot]com---I don't want to put contact information for other people in the comments.)

Are you in a leadership role in your school or district? Would you be in a position to shape grading and reporting practices? How are large-scale changes in your building or district usually handled?

It is an understatement to say that every school and district are different. Each one has its own culture, politics, and rules of diplomacy. If you're not a big fish, you can still do what's best for kids by implementing standards-based practices within your classroom---and letting it grow to other colleagues.

I wish I had more "how to" to share with you. Right now, I can only post about my own tools and classroom practices...my reflections on the process...and my dissertation research (on grading and student motivation). The implementation of things on a large scale is still a pretty new facet for most classrooms---I think a lot of us are feeling our way around.