I work and learn with several teams using rubrics to promote learning and growth. We have been working to translate our 4-point rubric scores to the 100-point scale required by our school.
It is that time of year. We want to report our learners’ progress to their parents, Grade Chairs, and other important members of their learning teams. While we understand the 4-point rubric score and what it means in terms of a child’s learning and growth, we feel that it is necessary to report their progress in more traditional terms.
We know that a single number can never represent the unique progress and learning of a child. We include a written comment with this number to provide additional information and evidence of learning. See the following blog posts for more information about these comments.
- Why I Do What I Do…
Quantum Progress, October 15, 2011
- Testing out some verbiage on quarter grades
Quantum Progress, October 15, 2011
- Grade Reporting: To comment or not to comment…that is the question!
Experiments in Learning by Doing, December 31, 2010
But, for now…We must have that single number.
We have worked together to develop a plan. We started by studying Classroom Assessment & Grading that Work and Transforming Classroom Grading by Robert J. Marzano. As a team, we have analyzed student work to calibrate our understanding of the rubric and how we score student work.
In Transforming Classroom Grading we read and studied Chapter 5. Assigning Final Topic Scores and Computing Grades and Appendix D: The Power Law Formula.
We investigated the following conversions by scoring student work and then analyzing the following scales to determine which scale most closely aligns with the team’s thinking about a score out of 100 points.
We looked at this data graphically. We wanted to see how a power function looked on the data.
Looking at Scale 1…
4 translates to 100, 3 translates to 90, 2 translates to 75, and 1 translates to 60.
It appears that a power function would fit the data.
It appears that this power function would over estimate the team’s rubric score of 2 when converted to the 100-point scale. Would there be another function that might fit better? Should we adjust the translations? We tried another type of function.
This is the same data – no adjustment in the translation – but we used a logistic model rather than a power function. Interesting, huh?
Looking at Scale 2…
4 translates to 100, 3 translates to 90, 2 translates to 75, and 1 translates to 65.
Looking at Scale 3…
4 translates to 100, 3 translates to 88, 2 translates to 73, and 1 translates to 65.
Numerically, the logistic function more closely converts our 4-point rubric scores to our agreed upon 100-point scale translation than the power function.
You are welcome to make a copy of our 4-Point Conversion E-PLC or 4-Point Conversion S-PLT Google spreadsheet and investigate for yourself.
This is where we are today. We have decided which of these scales works for our teams. We have calibrated our understanding and use of our rubrics. We have investigated these conversion tables numerically, graphically, and analytically. We have agreed to use the same conversion table to represent our learners’ work and progress.
This is a work in progress. We would love to know how you translate your rubric scores to the 100-point scale.
[…] please read this blog post by Jill Gough which explains the scale in more mathematical terms: https://jplgough.wordpress.com/2011/10/16/translating-rubric-scores-when-you-have-to/ Your students received/will receive a grade report print out this week. They also received an […]
[…] you squeeze by or fly over with a foot. The goal was to get over- everything else is just gravy. Jill Gough, math teacher and administrator, has written a wonderful piece explaining how we came to this translation number. Granted, it is only 4 possible grades out of […]
[…] How might we rethink grading and scaling? What does research tell us about translating scores between scales? […]
[…] Translating rubric scores when you have to… from Jill’s blog […]