All My Assessments are Pointless (But Not Purposeless)
I am not trying to say that my assessments are without purpose. They are quite purposeful, but my students are NOT assessed based on a collection of arbitrary points that have low reliability and often no validity. If you think about it...
- What makes something a 10-point problem vs. a 2-point problem?
- Does 1% of the total points = 1% of the total learning?
- Do points somehow represent the unit of learning?
Going pointless makes grades more meaningful!
Here is the rubric I use for each problem on any assessment that I give in AP Calculus AB:
Problem Rubric for AP Calculus AB
| ||
Grade
|
Description
|
Criteria
|
A
|
Mastery Demonstration of Objective
|
|
B
|
Proficient Demonstration of Objective
|
|
C
|
Partial Demonstration of Objective
|
|
D
|
Minimal Demonstration of Objective
|
|
After each problem is assessed and given a grade, the assessment grade is calculated using the following rubric. Notice that there is isn’t a point or percentage in sight.
Assessment Grade Rubric
| |
Grade
|
Criteria (based on each problem grade)
|
A
|
|
B
|
|
C
|
|
D
|
|
F
|
|
A breakdown of why points can lead to problems with giving fair and accurate grades:
- Multiple choice problems are often graded by points as all or nothing with rarely any regard to how well the students demonstrated mastery of a standard. If a student partially or mostly understands a standard, getting it wrong on a multiple choice problem does not show that. An easy fix is to grade each multiple choice problem is for students to include justification for why they chose what they chose. The justification can then be assessed by a rubric.
- As stated before, grading with points has extremely low inter-rater reliability. The grade for each problem based on points is influenced by who the grader is much more compared to following the Problem Rubric.
- An accumulation of points shows little validity in terms of grades communicating student demonstration of mastery. If a student scores 89% of the points, were they really just 1% off of full mastery?
- The arbitrary selection of how many points each problem is worth can make or break a grade. What would even dictate one problem being worth twice or three times as much as a different problem?
- Students can hide behind the averaging of points by skipping certain problems. This would show lack of demonstration of a standard, but not influence the grade enough to matter. Shouldn't everything on an assessment matter?
- Sometimes teachers give problems that are much easier than the standard requires as a grade boost to “help students out”. If grades are to be an accurate communication of what students are demonstrating, there should not be filler points added. This simply muddies the grade’s meaning.
- If you give points based on how many things a student can get wrong on a problem, that gives away a lot of information to students about the problem. For example, if a problem is worth 10 points, there might be more steps to complete it compared to a 2-point problem.
- Some teachers give extra credit points. If grades are supposed to be about learning, extra points make no sense.
- Averaging makes no sense when it comes to giving grades. If there is a whole in understanding of a particular standard, mastery of other standards should not be used to mask that.
Example breakdown of student grades with averaging and with the rubric:
What are your thoughts? Did I leave something out? Are there any positive reasons to use points besides teacher ease?
Let me know!