Spurred by part of a PLC discussion, I have been reading through our district’s document, Best Practices for Using Rubrics to Determine Grades. We currently formatively and summatively assess student performance using our Wind and Percussion Rubrics. Each criterion (tone quality, technique, etc.) has four different levels of performance (currently: exceeds standard, meets standard, making progress, not making progress). Our PLC discussion sprung from the completion of our 6-Week Assessment, transitioning into our 12-Week Assessment, and some professional development occurring at one of our middle schools.
The big question for us is: does the rubric accurately portray what a student knows/is able to do? Follow-up questions ensue like:
- Are we artificially inflating/deflating students’ grades? Is this because of the rubric/assessment?
- Is it reasonable for a student in X grade to meet the standard on each assessment?
Take-Away #1: New Descriptors of Performance Level
The document recommends we use the following for our different levels of performance:
- Exceeding the Standard(s)
- Meeting the Standard(s)
- Approaching Proficiency
- No Evidence – “Some piece of student work has been submitted, but it provides no evidence of meeting even a beginning performance level on any of the criteria.”
Take-Away #2: New Methodology for Converting Rubric to Grade
The document makes several points on pages 2 and 3 that really drove our conversation. I would highly recommend reading and processing on your own. To summarize:
- Students’ rubric level “scores” must be converted to grades based on the meaning of the performance levels. Do not directly convert rubric level scores to percentages (“points” earned divided by points possible).
- After marking on the rubric the descriptors that match the student performance, look across the criteria to judge the preponderance of evidence.
- Decide where the student evidence best fits with the performance level descriptors on the Summary Table below.
- Select the corresponding percentage within the performance level range to record in the grade book that is the best match given the evidence of student achievement against the defined criteria of the rubric.
- Give consideration to relative importance of criteria and the rubric level that represents proficiency.
- Determine the predominant performance level of the student work based on criteria in the rubric. Then use professional judgment to convert to the corresponding letter grade and/or percentage equivalent for recording in the grade book.
This table was the big spark for me. Like I mentioned above, we currently have 4 levels of performance. As a PLC, we have often discussed how the descriptors for each level of performance can sometimes box us in to a certain level of performance. Especially because some criterion have more than 1 descriptor, there is potential for a “high 3” or a “low 2.” Currently, we don’t have a method of dealing with that.
So how can we branch out from our current 4 levels of performance to something closer to the 14 described in our district’s document? How can we convert them in the way described in #2? This is a bit daunting considering we have 493 students being seen by 5 lesson teachers across 3 buildings. Is there a tool we could use like JumpRope or Google Sheets to better help us process this information?
Our Current Answer
And I don’t even really know if we can call it in answer yet. This is what we are planning to do to help us make a decision:
The 12-Week Assessment is due Wednesday, April 8th. For students in 6-8th grade, this means completion of a checklist of items as they prepare a solo for our Middle School Solo Festival on Saturday, April 18th. Students in 9-12th grade will be performing an etude in the next key signature in their method book (Student Instrumental Course Level 2 for 9th grade, Rubank Advanced Vol. 1 for 10-12th grade).
Each lesson teacher will grade the 9-12th grade etudes using the standard rubric via Google Drive that we have been using all year. Each teacher will also record a variety of samples they believe demonstrates a variety of performance levels.
After the conclusion of the Assessment Cycle, the team will listen to a variety of these samples and assess using the standard rubric to calibrate our scoring.
After scoring is calibrated, we will experiment with three different versions of the rubric:
- Old Rubric – More Options: Continues to use the 5 main criterion (breath support/tone quality, technique, rhythm, articulation, expression) with more levels of performance (4, 3+, 3, 3-, 2+, 2, 1, N/A).
- Expanded Rubric – Old Options: Expands the 5 main criterion based on each of the descriptors. Keeps the 4 levels of performance.
- Breath Support
- Tone Quality
- Register Changes
- Expanded Rubric – More Options: Uses the expanded criterion and more levels of performance.
We want to see how the different options of rubric effect the overall grade of the students’ assessments. Is it more/less accurate? Do we know more/less about what the students know/are able to do?