Rubrics are not a new concept in education. Their benefits have been touted for by many and a quick Internet search reveals large amounts of literature on the topic for anyone interested in learning more. Likewise, one does not have to look far to find examples of rubrics or websites that will create rubrics. Given this familiarity, it seems a bit surprising that they are still used incorrectly. While the benefits of rubrics are significant (see information under “Why use instructional rubrics” here), these benefits are not harnessed when they are used incorrectly.
There are two major misuses of rubrics that I see in action on a regular basis and I would like to take a moment to address them:
Rubrics are not assessment checklists.
Itemized lists with point values, like this, are not rubrics:
One of the dominant benefits of rubrics is that it forces the teacher to have a very clear understanding of exactly what the student did well and what they would need to do to improve and this information is communicated to the student. If a rubric is used well, the student will have a very clear understanding of what they need to do to improve in each category and will not have any need to ask the infamous, “why did I get this mark” question for any item on the rubric.
Rubrics, then, should have a descriptor for each level of achievement. On the above example, if neatness is a category on a five point scale (assessing neatness to determine if students know a social studies concept is a whole other topic for discussion), students should have a clear explanation of what category five neatness looks like, what category four neatness looks like, and so on.
In her article, “Using Rubrics to Promote Thinking and Learning”, Heidi Goodrich Andrade provides the following example of a rubric:
In contrast to the non-example rubric, Andrade’s rubric has clear explanations for each level of achievement. There is still some degree of subjectivity within the descriptors (as will typically be the case), but it is far more clear how each level of achievement is attained. Thus, the student knows what must be done to arrive at the next level of achievement for each category and can use this to inform his/her work from this point on. This is how rubrics become tools for learning.
2. Rubric grades and descriptors need to match
Rubrics are often written into a scale of four or five levels of achievement. Mathematically speaking, five levels of achievement would mean that each category on the rubric is calculated out of five. If percentage grades are given (another assessment issue that is yet another topic that needs to be addressed), that would mean that the categories compute to the following percentages:
- 100% (5/5)
- 80% (4/5)
- 60% (3/5)
- 40% (2/5)
- 20% (1/5)
The problem with this is that the descriptors for the rubric can often be generalized into something like this:
- 5/5 – great work. No improvement needed.
- 4/5 – really good work. Just a slight tweak needed somewhere.
- 3/5 – pretty good work. A bit of a misunderstanding or oversight is present.
- 2/5 – beginner stages. On the right track, but at the very beginning of the road.
- 1/5 – not there yet. Shows significant lack of understanding, or no evidence of understanding, or this component of the assignment was not included in the project.
The problem here is the disconnect between the descriptor and the grade. If a 2/5 reveals that a student is in the beginning stages, can we confidently assign a failing grade of 40%? It seems more logical to assign a grade that is at the lower end of the spectrum of a passing grade.
Just because something can be computed mathematically, does not mean that it should.
The lesson here is that, when designing a rubric, the numeric grades sometimes need to be skewed. Again, if reporting is done in percentages, the teacher needs to look at the descriptors of the rubric and determine an appropriate percentage equivalent that matches the descriptor, rather than assigning a percentage based on a mathematical computation. For example, a 90% might be more appropriate for a 4/5 on a rubric, rather than the computational score of 80%. Just because something can be computed mathematically, does not mean that it should. When using a rubric, it is often necessary to give each level of achievement a grade that is not mathematically derived.
As a side note, there really is no need to put an actual numeric score on a rubric. In fact, many would argue that that actually reduces the instructional component of a rubric. When a numeric grade is written on a rubric, students tend to look at the numeric grade and derive their sense of achievement from that, rather than having to read the descriptors to learn how they did and what they could do to improve. Putting a word descriptor for each level of achievement (rather than a 5-4-3-2-1) will be more informative for students and parents and will allow the teacher to assign a value for each level of achievement that better matches the descriptors than a grade out of five.
Rubrics can be incredibly effective tools to use to assess students, to teach them, and to show them where they are at and where they are heading, but it is necessary to ensure that our rubrics clearly communicate each level of achievement and are calculated in a manner that is fair and accurate.