Linda Suskie

  A Common Sense Approach to Assessment & Accreditation

Blog

Should rubrics be assignment-specific?

Posted on September 2, 2018 at 8:25 AM

In a recent guest post in Inside Higher Ed, “What Students See in Rubrics,” Denise Krane explained her dissatisfaction with rubrics, which can be boiled down to this statement toward the end of her post, “Ideally, rubrics are assignment specific.”


I don’t know where Denise got this idea, but it’s flat-out wrong. As I’ve mentioned in previous blog posts on rubrics, a couple of years ago I conducted a literature review for a chapter on rubric development that I wrote for the second edition of the Handbook of Measurement, Assessment, and Evaluation in Higher Education. The rubric experts I found (for example, Brookhart; Lane; Linn, Baker & Dunbar; and Messick) are unanimous in advocating what they call general rubrics over what they call task-specific rubrics: rubrics that assess achievement of the assignment’s learning outcomes rather than achievement of the task at hand.


Their reason is exactly what Denise advocates: we want students to focus on long-term, deep learning—in the case of writing, to develop the tools to, as Denise says, grapple with writing in general. Indeed, some experts such as Lane posit that one of the criteria of a valid rubric is its generalizability: it should tell you how well students can write (or think, or solve problems) across a range of tasks, not just the one being assessed. If you use a task-specific rubric, students will learn how to do that one task but not much more. If you use a general rubric, students will learn skills they can use in whole families of tasks.


To be fair, the experts also caution against general rubrics that are too general, such as one writing rubric used to assess student work in courses and programs across an entire college. Many experts (for example, Cooper, Freedman, Lane, and Lloyd-Jones) suggest developing rubrics for families of related assignments—perhaps one for academic writing in the humanities and another for business writing. This lets the rubric include discipline-specific nuances. For example, academic writing in the humanities is often expansive, while business writing must be succinct.


How do you move from a task-specific rubric to a general rubric? It’s all about the traits being assessed—those things listed on the left side of the rubric. Those things should be traits of the learning outcomes being assessed, not the assignment. So instead of listing each element of the assignment (I’ve seen rubrics that literally list “opening paragraph,” “second paragraph,” and so on ), list each key trait of the learning goals. When I taught writing, for example, my rubric included traits like focus, organization, and sentence structure.


Over the last few months I’ve worked with a lot of faculty on creating rubrics, and I’ve seen that moving from a task-specific to a general rubric can be remarkably difficult. One reason is that faculty want students to complete the assignment correctly: Did they provide three examples? Did they cite five sources? If this is important, I suggest making “Following directions” one of the learning outcomes of the assignment and including it as a trait assessed by the rubric. Then create a separate checklist of all the components of the assignment. Ask students to complete the checklist themselves before submitting the assignment. Also consider asking students to pair up and complete checklists for each other’s assignments.


To identify the other traits assessed by the rubric, ask yourself, “What does good writing/problem solving/critical thinking/presenting look like? Focus not on this assignment but on why you’re giving students the assignment. What you want them to learn from this assignment that they can use in subsequent courses or after they graduate?


Denise mentioned two other things about rubrics that I’d also like to address. She surveyed her students about their perceptions of rubrics, and one complaint was that faculty expectations vary from one professor to another. The problem here is lack of collaboration. Faculty teaching sections of the same course--or related courses--should collaborate on a common rubric that they all use to grade student work. This lets students work on the same important skill over and over again in varying course contexts and see connections in their learning. If one professor wants to emphasize something above and beyond the common rubric, fine. The common elements can be the top half of the rubric, and the professor-specific elements can be the bottom half.


Denise also mentioned that her rubric ran three pages, and she hated. I would too! Long rubrics focus on the trees rather than the forest of what we’re trying to help students learn. A shorter rubric (I recommend that rubrics fit on one page) focuses students on the most important things they’re supposed to be learning. If it frustrates you that your rubric doesn’t include everything you want to assess, keep in mind that no assessment can assess everything. Even a comprehensive final exam can’t ask every conceivable question. Just make sure that your rubric, like your exam, focuses on the most important things you want students to learn.


If you’re interested in a deeper dive into what I learned about rubrics, here are some of my past blog posts. My book chapter in the Handbook has the full citations of the authors I've mentioned here.

Is This a Rubric? 

Can Rubrics Impede Learning? 

Rubrics: Not Too Broad, Not Too Narrow 

What is a Good Rubric? 

What is a Rubric? 

Categories: Rubrics, Good assessment, How to assess

Post a Comment

Oops!

Oops, you forgot something.

Oops!

The words you entered did not match the given text. Please try again.

2 Comments

Reply ★ Owner
5:59 AM on September 4, 2018 
gasstation, thanks for taking the time to post your thoughts. One of the things I love about assessment is how quickly the conversation turns to teaching and curriculum design--which are what we should be focusing on! I'm an assessment person, not a pedagogy or curriculum expert, but let me give this a stab.

The challenge you're facing happens at lots of colleges, especially in entry-level courses at open-admission colleges such as community colleges. It sounds like you're engaging in individualized instruction, in which you set individual learning goals for each student depending on his or her needs and assessing achievement of those specific learning goals.

That's fine for terminal, elective courses (such as an art appreciation course taken by engineering students), but I worry about how it would work with courses teaching essential skills. To give you what I hope is a ridiculous example: you're teaching a college math class, and some students enter doing 2nd grade math. You and they work hard, and by the end of the course they're doing 8th grade math. This is an enormous improvement, but despite their great achievements, they're still not ready to tackle the kinds of math they'll see in subsequent college courses. Is it fair to pass them when they're not yet demonstrating at the level they'll need to succeed?

Again, I'm not an expert in this, but I think the answer is having your college rethink how it supports the underprepared students in your class. I know there's been really interesting research lately on how to provide this extra support, but as I said this is outside my bailiwick, and I'm not familiar with the details.
Reply gasstationwithoutpumps
12:39 AM on September 4, 2018 
I have a problem with "make sure that your rubric, like your exam, focuses on the most important things you want students to learn"?namely the assumptions that the most important things are the same for all students. My students have a very wide range of skills, and what is most important for one student to focus on can be completely different from what other students need. Trying to cover just the most important thing for each student would already result in a very long rubric (Note: I'm teaching classes of 80 students?it might be feasible in a class of 10 or 20.)