Header text

EssayTagger is a web-based tool to help teachers grade essays faster.
But it is not an auto-grader.

This blog will cover EssayTagger's latest feature updates as well as musings on
education, policy, innovation, and preserving teachers' sanity.

Thursday, March 1, 2012

Changing how we think about rubrics

Traditional rubrics are too general and macroscopic to help students. The future is specificity. And it's here.

Sharing rubrics is a simple, but important, way for teachers to collaborate.

Unfortunately, traditional rubrics -- by their nature -- can only address general, overall trends in a paper: "Some evidence was insufficient." That's fine for a quick, high-level diagnostic, but it's not very helpful for the student.

My goal when I'm grading papers is to coach the students so they can learn from their mistakes and do better next time. Traditional rubrics are good for setting expectations before the attempt, but once the essays are graded, they're really just an assessment tool. They're not a learning tool.

In order to improve, students need more fine-grained feedback: Which specific piece of evidence was weak? Why wasn't it compelling?

Traditional rubrics simply can't address these questions (nor, to be fair, were they meant to). Traditional rubrics are macroscopic. But students need the microscopic.

Specificity is king
So instead, let's create a new breed of rubrics that are designed for greater degrees of specificity -- going all the way down to specific, individual sentences. We should be able to evaluate each individual piece of evidence on its own, each transition, each body paragraph, each use of figurative language. And let's provide specific feedback to go along with each of those specific evaluations.

Showing a student exactly which pieces of evidence were solid and exactly which ones were weak is a lot more valuable than circling the "some" grid of a traditional rubric. Sure, we offer this sort of feedback when we write comments in the margins, but why do those comments exist separate from the rubric itself?

How about an example?
I had my seniors read an excerpt from "How a New Jobless Era Will Transform America" from The Atlantic. We discussed it on our moodle message boards for a few days and then I had them write a response piece to an audience of their choosing (e.g. their parents, teachers, peers). Here's my rubric. It has sentence-level, paragraph-level, and whole document rubric elements:

You can see the specificity in the sentence-level and paragraph-level rubric elements. It doesn't have the traditional rubric's general descriptors for each quality level. Instead it consists of specific feedback comments which are organized within the structure of the traditional rubric grid. And most grid squares contain more than one comment; problems with thesis or evidence come in many forms!

When I'm grading I'll use the "Article quote" rubric element multiple times per paper to evaluate each attempt the student made to incorporate evidence. And I'll select the feedback comment that is most relevant to each attempt (or add new comments to the grid as needed).

You can't do this with a traditional rubric.

And there's a good reason why: it's very hard to do this effectively on paper.

A little technology to the rescue
What is difficult on paper turns out to be super-simple with the web-based drag-and-drop interface I developed for EssayTagger. All it took is a little ingenuity, an English teacher with a Computer Science degree (me!), and 30,000+ lines of code.

This super-specific approach to grading may sound like a lot more work but my drag-and-drop interface with its reusable comments is a heckuva lot faster than scribbling the same comments in the margins over and over again.

And when I'm done grading I'll end up with some insanely cool and powerful data about my students' performance that otherwise wouldn't be available from traditional rubrics.

Traditional rubrics are dinosaurs and they do not do enough to help students improve. Specificity -- with some help from technology -- is the future.

Share your rubrics and discuss!
The link to my rubric that I posted above is a public link. No logins, no accounts necessary. All EssayTagger users can now share their rubrics, if they choose to do so (and you should!!).

I want teachers to be able to see EssayTagger's shared rubrics and to discuss this paradigm shift toward specificity -- regardless of whether or not they're EssayTagger users.

There's even a handy "print this!" button that'll format the rubric to look good in print (here's a look at the printable format) if you want to take it offline.

Teachers as publishers
We have essentially enabled every teacher in our system to become their own rubric author and publisher. You create the basic structure for a rubric (define the required elements and specify the possible quality levels), grade your essays and add comments to that rubric as needed, then publish your resulting rubric and share it with just your colleagues or share it with the whole world wide web.

Share it privately via email, post a link on your blog, tweet it, whatever you like.

The whole rubric creation system is built around what teachers do anyway -- comment on papers -- but it's all streamlined to be faster than traditional grading. And now you can share your work with essentially zero extra effort. All that brainpower and effort that goes into grading and providing great feedback is now being automatically preserved and can be made available to others.

That's freakin' awesome.

Import shared rubrics into your own assignments!
EssayTagger users have the added benefit of being able to import any publicly-available shared rubric into their own assignments. You can import a rubric and then customize it and its feedback comments however you like.

EssayTagger users: see the full step-by-step instructions for rubric sharing and rubric import!