Grading the Writing Process

My department team was planning our standard alignment plans - basically figuring out what standards from the Common Core we'd cover each year and each semester a student is in high school.  The rest of my department (which is filled with brilliant educators who are ten times better at their job than I am, so this isn't a jab at them, just a disagreement) felt that the writing process (CCSS.ELA-LITERACY.W.9-10.5) isn't a standard that should be graded, but discussed frequently with students.

And I'm all for discussing.  And showing them examples of my own writing that went through 10 drafts.  And showing them examples of great published writers who went through 20 drafts.  And probably doing some other stuff that showcases the awesomeness of the writing process.

But I'm also for grading it.  And I don't think it's a hard thing to do.

As it stands now (as this isn't refined or anything), a student who turns in an essay needs to do four things when they turn in their final paper in order to get a passing grade:

1. Turn in their prewriting, in whatever form it was, with their final draft.
2. Turn in their "zero draft," their initial attempt at starting/outlining their rough draft, with their final draft.
3. Turn in their rough draft with the final draft.  The rough draft must be a complete draft - no first-half-of-the-paper rough draft.
4. On their final draft, highlight all the changes between their rough and final.  (I stole this idea from Linda Christensen.)  Revisions must be significant - this isn't editing for spelling and grammar, it's revising.

On number four I "help" them out some (they would say "make them do extra work") - I make them write three different kinds of hooks and three different types of conclusions, then choose the best one.  Next year, I imagine I could do it with all sorts of different things.  Rewrite your thesis.  Reorder your introductory paragraph, trying the thesis statement at the beginning and the end.  Reorder your paragraphs in another logical way.

To exceed on this standard, students have to schedule a meeting with one of our peer writing tutors and show the changes they made after meeting with them.

I also let student revise their essays, often multiple times, which could be a grading nightmare, except for two things students need to do before I will grade their revisions:

1. Turn in your original graded rubric returned to you (I scan my stack of graded rubrics and save the final, because about half of the students will lose their graded one).
2. Highlight all the changes you've made between your final and the revised paper to be graded.  Turn in everything you turned in before.

Then I just look at what they added and determine if it addresses the concerns recorded on the original rubric.

But maybe this is foolish and silly.  Please correct me in the comments.

Chappuis, Chapter 3: Effective Feedback

In Chapter 2, Chappuis writes on how to help students get a clear picture of what they should be learning and how that learning should be demonstrated.  In Chapter 3, Chappuis offers five characteristics of effective feedback to keep students on the right path .

1. Effective feedback directs attention to the intended learning


There are two types of feedback for Chappuis: success and intervention.  These are straightforward.  Success feedback points out what a student has done well while intervention feedback tells them what needs to be corrected or improved.

There are common comments that do not accomplish the goal of effective feedback however.  Writing "incomplete" or "re-do" or really any grade does not encourage a student to succeed or give them a clear direction to go in.  As Chappuis (2009) states, "Providing feedback can be a labor-intensive proposition.  If we put all that time in we want to make sure that (1) we're doing it right, and (2) students will use it."  How many more students would have corrected an assignment if I had given them one-on-one feedback instead of writing "incomplete" at the top of their paper?  Even "see me," I would argue, has such a negative connotation, it is ineffective.  How many of your students actually do see you afterwards?

Often it is easy to praise students.  "You're so smart!"  However, Chappuis notes, this puts the feedback towards the learner rather than towards the work, and what does that make the student who receives the intervention feedback?  A study found that "look how hard you tried" is a much more effective comment because students see effort as something within their control, while "look how smart you are" is seen as a personal attribute and unchangeable (Blackwell, Trzensniewski, and Dweck, 2007, as cited in Chappuis, 2009).

Giving grades as part of feedback can be damaging as well.  Louann Reed taught at CSU that there are three terms used for assessment that shouldn't be interchangeable.  There's "assessment," the broad term; "feeback," which is written or verbal suggestions given to a student to help them improve their work or let them know their on the right track; then there are "grades," which contain a value judgement of the work.  The reason many students probably go to the last page of their essays for the grade and ignore the comments is because they see the feedback as unhelpful - why take the extra effort to make revisions when the grade's already been given.  Even if we say "you can make edits and turn this in for a better grade," how often does that actually happen?  Better to give the feedback, have them make changes, then assign a grade.

2. Feedback occurs during learning


Chappuis points out that we teachers often tell students "it's okay to make mistakes," because that's often how we learn.  By making a mistake, we know what not to do.  But sometimes our grading policies encourage the opposite viewpoint.  When quizzes and participation points are given leading up to the big test, how fair is it to grade them both.  It's double jeopardy.

Louann's continuum also should go in order.  Feedback should always come before grading.  Feedback should happen during learning.

3. Feedback addresses partial understanding


Feedback should be given when it can help a student move forward.  To give a student who needs further instruction feedback would be more frustrating to both student and teacher than helpful.  Only when there is partial understanding is feedback helpful.

4. Effective feedback does not do the thinking for the student


The best example I have for this is correcting conventions in an essay.  If one has taught sentence fragments, comma splices, and the use of semi-colons, rather than correcting each fragment in completed essays, mark the line where the error occurs and have the student correct it.  Students have a task and incentive to use the teacher's comments/feedback to move their learning forward.

5. Effective feedback limits corrective to what students can act on


Differentiate feedback to what an individual student can handle at one point in time.

Methods to offer feedback

Chappuis (2000) suggests using picture and symbol clues (p. 75) like stars and stairs to make comments on student work.  Students could also offer their own stars and stairs - what they think they do well and what they need to improve - before the teacher adds their comments.  Two-color highlighting (p. 82) is another way for students to self assess.  They highlight the sections of the rubric they believe they meet before the teacher, using a different color highlights their own assessment.  By using two primary colors, this could easily identify for the student where they and the teacher are in agreement.

Chappuis, Chapter 2: Where Are We Going?

In Chapter 1 of Seven Strategies of Assessment for Learning, Chappuis presents readers with her seven strategies:

1. Provide students with a clear and understandable vision of the learning target.
2. Use examples and models of strong and weak work.
3. Offer regular, descriptive feedback.
4. Teach students to self-assess and set goals.
5. Design lessons to focus on one learning target or aspect of quality at a time.
6. Teach students focused revision.
7. Engage students in self-reflection, and let them keep track of and share their learning. (Chappuis, 2009, p. 12)

In Chapter 2, Chappuis begins moving students to meaningful self assessment by establishing clear learning goals for them to pursue. She points out that strategies three through seven rely heavily on the first two - without successful implementation of the first two strategies, the rest will not be successful either.

Provide students with a clear vision of the learning target

For the first strategy, Chappuis (2009) gives three steps (p. 22).

1. Share the learning target with students

Many teachers I know, including myself, post learning targets for their students on a whiteboard/blackboard or towards the beginning of class with a digital projector. Yet often we make the mistake (myself included) of using eduspeak:

Today's Learning Goals:

• EL.08.RE.05 Match reading to purpose--location of information, full comprehension, and personal enjoyment.

As soon as the student hits EL.08.RE.05, their attention turns elsewhere. Not to mention the whole "match reading to purpose" bit; I didn't learn what that meant until college, despite the fact I'd been doing it since I started reading.

2. Use language students understand

Instead, it would be much better to say:

Today we will learn:

• How to determine a goal or purpose for reading a text: looking for information, understanding a point of view, or for entertainment.

Chappuis has some pointers when rephrasing a learning target. She suggests stating the goal within the sentence structure "I am learning to _____________" or "we are learning to ___________." She also states that teachers should identify words or phrases that need clarification for their age group and use a dictionary definition to help decide the best way to simplify them.

3. Introduce students to the language and concepts of the rubrics you use

Chappuis encourages the use of rubrics for all the learning goals. Taken at face value, this could seem painstaking at best and insurmountable at worst. However, Chappuis stresses that the rubric should be both general and descriptive; they should be applicable to different assignments and use "language that explains characteristics of . . . performance at increasing levels of quality" (p. 40).

In the past few years in the blogosphere, at least, rubics have gotten a bad rap. I have at least a couple unpublished drafts that defend the rubic, but I'm forced to make their case here. Good.

The firestorm started shortly after Alfie Kohn published "The Trouble with Rubrics" in the March 2006 edition of English Journal. Kohn correctly admonishes teachers who use rubrics solely because they "make assessing student work quick and efficient, and they help teachers to justify to parents and others the grades that they assign to students" (Andrade 2000 as cited in Kohn 2006). There's plenty wrong with that statement that one can read about in Kohn's article.

He continues by stating that rubrics should only make up one piece of assessment because they too exactly spell out just what students need to do in order to squeak by, stifling the pursuit of learning. This, Kohn (2006) argues, is why "all bets are off if students are given the rubrics and asked to navigate by them" (p. 13), author's italics. I agree that this is a common symptom of students in our schools, but I doubt rubrics are the sole culprit. Any form of grading system that tells students they need to receive so many points in order to receive n grade is asking for trouble. "Smart" students care about grades. The "dumb" ones are smart enough to know that grades don't really matter - it's whether what they're learning is relevant enough and they are proficient enough to survive in society.

Rubrics break down complex processes that take years to master into bite size chunks that help students understand how they can improve. Grading writing (or anything) holistically is acceptable only summatively, and for me that means at the end of the year.

The best, most understandable means of formative assessment for students is that presented in rubric form. As Kohn (2006) says, "it matters whether the objective is to (1) rank kids against one another, (2) provide an extrinsic inducement for them to try harder, or (3) offer feedback that will help them become more adept at, and excited about, what they’re doing" (p. 14).

It would be interesting to get Kohn and Chappuis in a room together; Kohn (2006) writes "any form of assessment that encourages students to keep asking, 'How am I doing?' is likely to change how they look at themselves and at what they’re learning, usually for the worse" (p. 14), my italics. I'll reserve final judgment until I finish Seven Strategies, but in my mind Chappuis is winning.

Her rubrics seem to deal with many of the causes of Kohn's concern. The rubric should avoid evaluative ("excellent organization of thoughts") or quantitative ("thoughts are organized into three paragraphs") language and focus on descriptive ("the writing is organized using any of the various conventions of language [punctuation, sentence fluency, paragraph breaks, headings] to make thoughts easily understandable to the reader") and diagnostic language. Evaluative and quantitative language focus on the work, while descriptive language focuses on the learning.

Use examples of strong and weak work

The creation of rubrics moves fluidly into Chappuis' second strategy.

Showing students examples of strong and weak work allow them to compare their own work to a final product. In addition, Chappuis suggests having students choose weak and strong examples from a number of examples and state why one is strong or weak, either with a set rubric or as a precursor to creating a rubric together as a class.

Chappuis also emphasizes how important it is that the work shown to students is not from their class. In the past I have shown work from students in the classroom, thinking they would not reveal their identity. Unfortunately, this was not always the case. More than once, a student's first response to seeing their work on the screen is “hey, that's mine!” Whatever lengths the teacher may go towards making the work anonymous can be unknowingly sabotaged by the very student the teacher is trying to protect.

Instead, Chappuis suggests choosing or creating at least three samples for each trait on a rubric from other classes or previous years – a strong, mid-range, and weak example. Give these samples to the students one at a time either on the overhead or, better yet, as individual copies. Have them Think-Pair-Share. First, students individually decide how they would score the product on the rubric; students decide if the work is predominately strong or predominately weak. If strong, students start on the high end of the rubric moving down, if weak, the opposite, before settling on a score that meets most of the criteria. They then discuss with a partner, using the language in the rubric. The class then votes and gives reasons why they chose the scores they did before the teacher reveals the score they gave the example. The process is then repeated for each remaining example or until students understand the language and concepts in the rubric, whichever comes last.

This would be an excellent way to teach 6-trait writing. I attempted it last year, but stopped when I ran out of examples but before students fully understood the concept. Hopefully this year I have more samples to draw upon.

Works Cited

Chappuis, J. (2009). Seven strategies of assessment for learning. Portland, OR: Educational Testing Service.

Kohn, A. (2006). Speaking my mind: The trouble with rubrics. English Journal, 95(4), 12-15.