31

I teach a practical class which is assessed by a lab report. I grade the lab reports based on a rubric. The rubric has 6 sections (abstract, introduction, methods, results, discussion, conclusions) with a number of items that I generally expect to see in each section. Historically, about 95% of the lab reports can be accurately graded based on the rubric while the remaining 5% of the lab reports go in unique, often very good, directions and therefore do not tick many of the boxes on the rubric. I am considering showing the students the grading rubric in advance of writing the reports this year with the hope that this will key them into what is important so that they can better demonstrate their understanding of the key issues. Is there any research that looks at the benefits and consequences of showing students a grading rubric in advance?

7
  • 1
    I have only seen rubrics which are distributed with assignment briefs. The rubrics, however, do not provide a step-by-step. That is, no student would read a rubric and know what they should write (just how they should write it). For example, 70%+ (UK) requires work to be free from spelling and grammar errors (as one simple example). I would not include anything in the rubric which would limit their creativity.
    – earthling
    Commented Oct 26, 2014 at 22:44
  • 2
    Probably not the case here, but in many situations releasing a detailed rubric can be self-defeating; often a large part of an assignment is researching the subject matter and deciding what best to focus on.
    – sapi
    Commented Oct 27, 2014 at 8:10
  • 1
    I should be clearer than I was above. Rubrics should not include any task-specific information (which means rubrics should be able to be reused across subjects).
    – earthling
    Commented Oct 27, 2014 at 11:57
  • 7
    But isn't it more fun to play "Hide-the-ball" with the students, and then arbitrarily punish those who did the worst job of reading your mind? +1 for raising the issue.
    – Aaron Hall
    Commented Oct 27, 2014 at 17:57
  • @AaronHall - not sure how to take your comment, so I'll simply say that having unstated requirements and rewarding "successful" mindreading stands a fair chance of demotivating the majority of (students/workers/whatever). IMO the general lab report outline must be clearly stated if students will be graded on their adherence to it. Commented Oct 27, 2014 at 23:46

6 Answers 6

24

Being open and clear about your grading policy is a large part of what makes the grades you assign meaningful. In that respect, it is generally helpful to share the rubric with your students if you work from one. At minimum, a good grading system should meet three criteria:

  1. it should accurately reflect differences in student performance
  2. it should be clear to students so they can chart their own progress
  3. it should be fair

Sharing your rubric directly enhances the second criterion, and presents a context for evaluating the other two. You might also consider using some of your lecture time to go over examples of what an excellent lab report should look like and then discuss the rubric with your students. This would give you an opportunity to point out that lab reports should cover key points, but good lab reports don't necessarily follow a rigid cookie-cutter format.


Since you asked about sources on sharing rubrics with students:

The authors of Introduction to rubrics, mention discussing the grading rubric with students and even include a chapter on constructing/tailoring rubrics directly with student feedback.

[Stevens, et al.] "... because we discuss the rubric and thereby the grading criteria in class, the student has a much better idea of what these details mean ..."

However, you may be interested in this paper which examines the learning outcomes for different peer groups when they're given details about assessment criteria. Evidence for their conclusions is based on a very limited number of samples, so the usual cautions apply, but they found that simply sharing explicit grading criteria was not sufficient to positively influence learning, while making time for the students to work more intensively with the rubric did yield benefits.

[Rust, et al.] "... it is being engaged with the process of marking as well as seeing examples of other work that significantly contributes to the students’ subsequent improvement in performance."

1
  • +1 for sources. The latter seems to imply students benefit from the rubric actually being part of the pedagogy, instead of just metadata.
    – abathur
    Commented Oct 27, 2014 at 14:59
49

Sharing a good rubric with your students can be a helpful way to let them understand what is expected of them. There is a down side in that the rubric then provides a basis for students to complain that you've graded a paper unfairly. Don't release the rubric unless you're really willing to give credit according to the rubric even if a paper has obvious flaws in areas not covered by the rubric.

7
  • 8
    I agree with the other posters. If you are having problems with rubrics, instead of keeping them secret, fix the rubric. For example, if you get some lab reports that are unique, but do not follow the directions, then put something in the rubric that allows you to reward creativity. Commented Oct 26, 2014 at 20:36
  • 2
    I agree that having a good rubric is the best solution to this problem- the difficulty (and the author of the question mentions this) is in writing a rubric that really covers the wide range of possible correct and incorrect student responses to the assignment. In my experience, rubrics work reasonably well for things like term papers but not as well for the kinds of problem sets and exam problems that are common in undergraduate engineering and science courses. Commented Oct 26, 2014 at 20:56
  • 3
    If you are worried that releasing a rubric will cause students to think you are unfair, it sure sounds like either your rubric is a poor description of the way you are actually grading and should be fixed or that you are actually are being arbitrary and unfair.
    – mako
    Commented Oct 26, 2014 at 21:42
  • 3
    As an example, of this, I once gave an exam question that required the exam taker to consider a mathematical proposition and either prove it or provide a counterexample to show that the proposition was false. The student answered with both a (incorrect) proof and a valid counterexample. My rubric didn't account for that possibility- it would have given the student partial credit for whatever part of the proof was OK, and it would have given full credit for the counterexample. Commented Oct 26, 2014 at 22:55
  • 3
    @BrianBorchers: I agree. Never underestimate a student's ability to come up with an answer beyond the scope of your rubric.
    – Paul
    Commented Oct 27, 2014 at 19:31
8

There are disadvantages to distributing the rubric. Most importantly: it encourages sensible students to game the rubric rather than understanding and answering the question. I do not think this concentration on grading rather than learning is a good thing. Against this you must set the benefits of transparency and the ability of students to engage with their assessment.

You seem to think that the very good unique answers are a problem to be discouraged – I disagree; I think they’re a good thing. The difficulty for you is in grading them – and it seems you are sticking to your rubric in the face of good answers. This is a mistake. The rubric should be used as a guide to help you give consistent grades, not to limit the range of answers you will accept. When faced with a high-quality answer that does not fall neatly into your pre-written rubric, your response should be to try and grade that submission without the rubric not reduce the marks you award.

0
6

Write careful and comprehensive rubrics and distribute them with your assignments. What better way to tell students what you expect from the assignment than by telling them exactly how you are going to evaluate it?

If there are unique and interesting directions that you want to allow assignments to go but that you think your current rubric precludes, rewrite your rubric so that it's flexible enough to allow these papers to be assessed highly. A great rubric is clear enough that an instructor can communicate clearly what is expected in a way that is transparent, fair, and clear, but not so overconstrained that it leads students to formulaic box-checking. It's a tricky balance but worth striving for.

2

I wanted to continue the trend of substituting my opinion for the research you requested:

Any impact the sharing of the rubric itself has on student outcomes is secondary to the impact the transparency of sharing the rubric will have on you, your pedagogy, and the course itself. Even a controlled study of a large standardized course taught in many sections, half of which do/don't receive rubrics, is going to be blind to the cumulative benefits of transparency:

  • being forced to clarify your expectations to yourself (i.e., extract them from your intuition) well enough create a rubric you're confident leaving in the hands of your students
  • being encouraged to evaluate how well your pedagogical practices actually prepare students to meet these explicit expectations
  • being forced to face problems with your expectations (as the distance between the grade you intuitively want to give a student and the grade the rubric suggests they deserve)
  • being forced to renegotiate your rubric as students challenge it

In short, your ability to consistently apply your public rubric in ways that satisfy both you and your students is a good proxy for evaluating your own course design.

Beyond this, I'd also like to add to the "student benefits" concern: one of the overarching goals of education is to teach people how to do "good" work on their own. Learning to self-evaluate is a big part of consistently performing good work without supervision; learning the qualities of good work is a big part of learning to self-evaluate, along with feedback mechanisms that consistently communicate the importance/validity of these qualities.

1
  • 1
    Ha eight years later and this opening line is still making people laugh. Good point that no one is actually posting research.
    – JimLohse
    Commented Mar 11, 2022 at 17:26
1

In my training to be a public school teacher mostly from principals, assistant principals and other expert teachers, the expectation is we show the rubric when we give the original assignment.

In a related example, my college tutoring student with a long test has likely spent too much time on less critical questions because the professor did not explain the weight given to each question. Even when he got his test back he and I are unsure how much weight is being given to each question. This is inappropriate on the professor's part.

It is only fair to students that they know ahead of time what is important to the professor communicated through a rubric or advance knowledge of the grading scheme.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .