When discussing grading with other physics teachers, the one thing that I always emphasize the most is the value of getting rid of points. When I switched to standards based grading two years ago, there were many benefits, but dropping points is the one that I celebrate the most. The decision to assess without points is one of the best and most freeing choices that I have made in my nine-year teaching career. I realize now that much--far too much-- time was spent fussing over points with very little benefit to student learning or assessment accuracy.
Right now you may be thinking that points are the only way to assess fairly and accurately. If you've used points for your entire career, it might seem impossible to do otherwise. Trust me--it's not. In humanities classes it's very common to assess student work using rubrics (without points). When humanities teachers grade a student's essay with a rubric, they are looking to assess a certain set of skills. This is not so different from what we're doing when we grade a physics test. It stands to reason that we should be able to assess student work in a similar manner.
Just to be clear, you will still need to use a point, percentage, or letter grade system in your gradebook (unless of course your school uses some sort of standards-based reporting system). I'm talking about getting rid of points as a means for how you arrive at those scores. You may still have assignments in your grade book that are worth ten points. What I'm saying is that giving a score like 8/10 shouldn't mean that a student earned eight out of ten points on the task. We need to get away from the "you earned x out of y possible points" mindset.
Objectivity and Accuracy Are Not The Same.
Hiding Behind Points
Up until I started using SBG, I graded tests using a points scheme. MCQs were each worth a certain number of points and FRQs were assessed using scoring guides that were created in a fashion mimicking the AP scoring guides. While assessing tests, my aim was always to be as impartial as possible. Like many instructors, I graded tests one page at a time, often not even knowing whose work I was seeing. It seemed like the most objective (and therefore the best) approach.
I now realize that I was hiding behind points as a way of creating an illusion of objectivity. Points were used as a tool for rationalizing a certain grade. But remember, its not the scoring guide or the questions that we really care about! What we want to assess are a student's skills in relation to particular achievement standards. Rather than using points as a sort of "middleman" why not go straight to the standard or the skill that you are hoping to assess?
As teachers we use scoring guides and point systems as a way of removing teacher ownership from the assessment process. If a student disagrees with the score they earned in a points-based system it's easy for instructors to point to the scoring guide and make a clear case about why they earned X points on a given question. It completely circumvents the question of our own professional judgement. We don't have to explain why we decided to award a particular score. It's out of our hands, we followed the rules when we tallied up their points, and the student earned what they earned. This approach is cowardly. Teachers should be able to justify student scores in relation to performance expectations. Having points to hide behind is not a good enough justification for using them.
Also, don't forget that the teacher was the architect of that "objective" grading system to begin with--a system which is almost certainly imperfect and will unintentionally favor some students over others as a matter of luck or circumstance. Just because a system is objective, that doesn't mean it leads to accurate measurement of student learning. As an analogy, consider a perfectly functioning justice system. Let's say (and I know this isn't true) that all existing laws were enforced equally among all populations in the US. Does this mean that the justice system works? Not necessarily! Sometimes there are extenuating circumstances that should be accounted for during sentencing. Sometimes a new situation arises where the law doesn't seem right and a more sophisticated judgement needs to be made. I hope I'm getting the idea across!
Scenarios Where Points Fail
To drive this notion home, let's take a look at some scenarios where the "objectivity" of points fails to result in accurate measurement of student learning. You have probably found yourself in some of these situations before:
A student showed clear evidence of understanding through their approach on a given question, but the way you constructed the scoring guide didn't allow you to award as many points as you know their work merits.
A student accidentally stumbled into a correct (or partially correct) answer that is clearly based on flawed reasoning or conceptual misunderstanding. The scoring guide results in them earning some points essentially through good luck.
You wrote a test question that was really inspired--one you're really proud of. It was creative and required a unique transfer of skills or content. However, there was a bit of twist that many students struggled to see. As a result many competent students bombed the question. Other students, happened to see the "twist," and did very well. This resulted in a big point disparity between two groups of students although you know that there was not actually a major difference in skill between those two groups.
A student finished 3/4 of the test and left the last 1/4 blank. Their work on the portion that they finished is outstanding. It is abundantly clear that they have mastered many of the standards that you were trying to assess. You tally up their points and walk away knowing that they didn't come away with a score that really reflects their ability.
Bottom line: We are professionals and we know the different skills students need to master. We can define criteria for what different proficiency levels look like. Rather than hiding behind the illusion of objectivity provided by points we should be using our expertise to make informed judgements about where student performance lies along a proficiency scale. If you have the courage to jettison points you will not be beholden to a scoring system in situations like these. Instead you can allow yourself the flexibility to assess students in a manner that you know to be more accurate.
Points Highlight Differences in Performance That Don't Actually Exist.
When I used a traditional points system to grade tests, I would tally up points, apply a curve, and the convert the final score into a % out of 100. Let's consider two hypothetical students who are assessed within this system. Two students get back a test. One earns a 79% and the other receives an 81%. Both of these scores will have similar effects on their overall grades and as teachers, we know that there was probably not much difference between the performance of these two students on the test; honestly, it's probably all but indistinguishable. From a student's perspective however, there is a world of difference.
A score in the 80's feels very different from a score in the 70's. Depending on what grading scheme you are using, it may be the difference between a B- and a C+. Both students (and parents for that matter) tend to react very differently to scores that appear in different grade ranges. Taken to a different extreme the difference between a 59 and a 60 can be the difference between a failing and a passing grade! Should such a significant judgement really come down to a single point?
When a student earns a grade there is certainly a margin of error that goes along with it. This margin of error is a result of many small factors. Things like: the way we worded a question, the way we chose to draw up the scoring guide, the way the student was feeling on test day, the questions that we chose to put on the test, etc... We don't know what a typical margin of error is because there isn't a clear way of determining what it is. If we give a grade of 87% we don't include error bars, but there is uncertainty in this measurement. We just don't know how much uncertainty. Should it be an 87±1% or 87±5%? It depends on the quality of the assessment, how many opportunities were provided to show a skill, etc... My guess is that in many situations the uncertainty is much closer to ±10%.
When you grade a test and give a final score using points, you are reporting out a score that reflects student performance with greater precision than you actually know. Students and parents can read too far into these small differences and this can have profound effect on a student‘s self-confidence and emotional well-being.
What About Student and Parent Pushback?
Before I rolled out this grading scheme to students I was really worried about this. I teach at a school where students (and parents) can be extremely concerned about grades and any perceived unfairness. I thought students would view this system as too subjective.
Anticipating this challenge, I addressed this upfront from the very beginning of the year. I reminded students that their tests were never graded in a vacuum and that I would always ensure that they were being graded alongside one another so that no one received unfair treatment. (I also introduced my very generous reassessment policy at the same time which I'm sure helped as well.)
There were one or two concerned students, but overall received far less pushback than I had predicted. In fact, when I scored students using points, I would often have students come in to squabble with me over single points or half-points that were deducted on test questions. Without points such minutia aren't really a concern because even if I made a small mistake marking a question, it likely wasn't significant enough to have affected their final score.
Assessing Without Points Encourages Teachers to Look at Student Work with The Right Mindset.
Getting rid of points is not only a change in reporting but also a change in the mindset of how you look at student work.
As I mentioned before, I used to grade tests one page at a time to make sure I did it the same way for every student. Much of my time was spent making decisions like deciding whether a student should earn 4/7 or 5/7 on a certain FRQ. All of my energy was spent on trying to be fair. It was taxing and yielded very little true value for students.
Now when I grade, my energy is spent on really seeing the student’s work fully. Not losing sight of the forest for of the trees (or something like that.)
Grading is about examining evidence of achievement for each student. It is about seeking a comprehensive picture of where a student stands in relation to the skills I want them to learn. I now take a student's test and look at it holistically for demonstration of achievement on each skill that is being assessed. This has led to a much more personally rewarding experience when assessing students. It allows me to walk away from grading with a deeper understanding of each student and to develop a more complete picture of their growth throughout the course's trajectory. Now, when I sit down with students to go over their tests with them or to reteach concepts, I have a much better sense of their strengths and weaknesses. In many cases I can even remember specific mistakes that students made.
Of course, I still try to be fair. As I score each student I flip back to tests I've already scored to make sure that the scores are calibrated and that I'm not getting more generous or more strict as I become fatigued or impatient. The difference is that calibration is no longer the main focus of a grading session. Getting rid of points has completely changed the way that grading feels. Grading tests is no longer just a long slog; it's professionally rewarding.