Rubrics Aren’t What They Used To Be

Version 2For music teachers assessing students, rubrics are usually the most effective means. Rubrics can be used to assess pretty much anything. That’s good for music teachers, because we assess a host of items in our students’ performances, ranging from accuracy, to tone and overall musicianship. Aspects of music making that were once thought to be not assessable because they were to subjective can, we now realize, be assessed with a well constructed rubric. But the assessment capability of a rubric is only as good as the mearueablility that is built into it. Today, I’m going to look at what makes a good rubric.

When assessing with rubrics became popular in the 1990s, it was common for rubrics to lack specificity, making precise assessment problematic. For example, as an All-State adjudicator, I was given a rubric that, for note accuracy, gave me the following choices: “The student performs…all pitches/notes accurately, most pitches/notes accurately, many pitches/notes accurately, some pitches/notes accurately, few pitches/ notes accurately.” This vague approach to rubric construction is still with us today. Many of the Model Cornerstone Assessment rubrics are similarly unclear. There is no way of knowing what differentiates “most,” from “many,” or “some” from “few.” If more than one adjudicator assess a single performance, it is likely that each adjudicator will give a different assessment because of different ideas about what each of those terms, most, many, some and few” really mean. Any rubric that uses these kinds of terms cannot give a consistent or consistently accurate assessment of student work. Said another way, it is impossible for an adjudicator to defend why he or she decided on “many” instead of “most.” A useful rubric most be specific as to what the student did or did not do.

So how could we improve this note accuracy item on this rubric? The first choice, “all pitches/notes accurately,” is fine as is. There is only one way a student can receive this score, and that is to play every note right. One mistake bumps the assessment down to the next choice. That is where we must become more specific. The simplest solution is to specify a number of mistakes at each level. It is decidedly impractical for an adjudicator to count the number of correct pitches/notes played or sung, but it is realistic for an adjudicator to count the number of errors made. Instead of “most pitches/notes accurately,” we can use, “with 1-3 incorrect pitches/notes.” Giving a range of errors captures the spirit of “most” while providing concrete parameters to the description, and where more than one adjudicator is assessing, allows the same score to be given, even if one adjudicator overlooks on or two errors. The other choices would similarly be revised, so that “many pitches/notes accurately” becomes “with 4-5 incorrect pitches/notes,” “some pitches/notes accurately” becomes “with 5-6 incorrect pitches/notes,” and “few pitches/notes accurately” becomes “with more than 6 incorrect pitches/notes.” The numbers chosen can be different from the examples given. The point is to have a small range of errors that draws clear boundaries around each choice, and describes the performance in a way that is useful to students and teachers alike.

Here’s another example, this time one that addresses response to listening for the general music teacher. One of the objectives in the National Core Arts Standards is that students will “be able to  demonstrate and describe how a response to music can be informed by a personal context.” At first, this seems like it might be one of those items that cannot be assessed. How do you measure or know a person’s personal context, let alone demonstrate how it relates to a response to music. I have recently given this some thought as I worked with my Professional Learning Community on writing a 3rd grade instructional unit for listening. We had to think creatively to come up with a solution, and we won’t really know how well it works until we’ve piloted the unit, but here’s what we came with. We decided that we could know a person’s personal context based on four sources: the person’s description of themselves (who am I), description of their friends (with whom do I choose to develop relationships), hobbies they enjoy (what interests me), and cultural background (what I have experienced in life). Students can describe these things by talking about them or writing about them, and they can demonstrate them by interpreting music or examples of music in a way that expresses each item. Thinking of music in terms of self-identity gives students the chance to thoughtfully and intentionally shape their musical world according to how they see themselves and their place amid the people and places that form their enviroment. 

So how does a rubric look that assesses a student’s ability to “demonstrate and describe how a response to music an be informed by a personal context?” One solution is given below. This assessment task was to assess students’ ability to select a musical work to which to respond, and to support their choice with evidence. This type of response and assessment correlates well with language arts common core, a definite strength for music teachers.

Criteria 4
(Does Not Meet Standard)
Students will be able to  demonstrate and describe how a response to music can be informed by a personal context. Described and demonstrated (through performance example) how the selected musical work relates to self-description, description of friends, hobbies, and cultural background. Described and demonstrated (through performance example) how the selected musical work relates to self-description, description of friends, and hobbies. Described how the selected musical work relates to self-description, description of friends, and hobbies. Selected a musical work, but did not describe how it relates to self-description, description of friends, or hobbies.

Notice that as the score becomes greater, the number of things the student includes in his or her responses also becomes greater. At the lowest end, the student has still made a response, they have selected a musical work to which to respond, but have not supported their choice in any of the ways being measured. Language such as “in some detail” and in great detail” have been avoided, because, like most notes and many notes, they don’t define clear differences between one level of performance and another.

Good rubrics make clear what learning objectives are, and what students have learned or are able to do as a result of instruction. They give teachers valuable feedback that guides them in planning further instruction, and students valuable feedback that guides their further learning. Rubrics that are not specific enough to do so should be revised or replaced with rubrics that plainly state measurable items.


Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s