Calculating final grades always brings up lots of philosophical-mathematical controversies. NY state tests use a 1-4 rubric, where 4 is "far above the standard," 3 is "above the standard," 2 is "below the standard," and 1 is "far below the standard." This has some advantages. First of all, for so many assignments, the difference between an 88 and a 92 is not great and doesn't communicate much about the work. Using a 1-4 rubric encourages teachers to explain
to students what is expected in order to meet or exceed a standard, and why certain assignments do not meet a standard. This is more useful than simply assigning a number grade. In most lines of work, adults do not get a score, they get feedback or judgment on the overall quality of the projects they complete. Also, using a rubric discourages cut-throat competition among students, without eliminating it altogether. It shifts the focus to producing high-quality work, not just getting the highest number.
Private high schools and other selective programs are still interested in regular percentage-based grades, though. An advantage of percentage grades is that parents, many students, and institutions understand them. They also provide for fine gradations in judging work; for example, getting an 88 may push a student to look for ways to improve in order to get into the 90s. A score of 3 on a 1-4 rubric doesn't give much information about whether the work solidly achieved the standard, or just barely squeaked by, or was close to being superlative, 4-level work. Also, percentage grades are more useful for grading traditional quizzes and tests, which are still an important facet of assessment.
Taking all this into consideration, we decided to give BOTH kinds of grades on our report cards this year. This posed several problems:
For my Communication Arts colleagues, who score all classroom assignments using the rubric, the question was how to make a broad category into a fair and meaningful percent? If two kids get a 3, that doesn't mean it's fair to give both kids an 85% (or some other number) and leave it at that, since one child may be achieving close to a 4, while the other is barely above a 2. So, they created a conversion chart. If the average of a child's homework grades came out to a 3.6, that translated directly to a percent. To my mind, there's no point to that, since it's the same number on a different scale; in what way does that broaden or deepen the information provided to the child, parent, or high school about the child's performance? I guess the only purpose is to make the number more familiar to people used to the percent scale.
For me, the problem was the meaning of a percent grade. I use a mix of systems in my grading; homework is done using the rubric, because homework is graded on many things, including neatness, completeness, and correct answers; quizzes and projects are graded using the percentage scale. Our school claims that our minimum passing grade is a 75%, below which you are put on academic probation and get weekly progress reports. So, I interpret a 75% as the floor of the 3 category - that's our standard, and if you get 75% or higher, you met the standard. I interpret anything from 90-100% as a 4 - you're doing some seriously high-quality work. And a 2 is anything between a 65%-74%, since most schools use a 65% as the minimum passing grade, so that seems like a good place to start the "almost meeting the standard" category. Below 65% is a 1, of course. As you can see, the rubric does not translate mathematically to the percent grade, but it communicates additional information about how I think the child is doing and what I consider acceptable work. Other teachers in my school thought that 100% should equal a 4 and 75% should equal a 2.5, and everything else be calculated mathematically from there.
Anyway, it gets even more complicated. When my computer grading program calculates a student's average homework grade, it makes a 4 into a 100%, 3 into 75%, 2 into 50%, and 1 into 25%. So, although I say
that to me a 3 is a 75%-89%, it really isn't! This method brings down the grades, compared to their meanings in my mind.
The bottom line is that when I calculate my grades, I feel comfortable with the results. I drop the lowest homework grade every marking period, to leave room for one mistake or bad day. It's enough to make a difference for a few kids, even though I have more than 25 homeworks per student. This marking period, I also dropped the lowest quiz grade, to allow for adjustment to the format and requirements of the quizzes. The grades I give line up really well with what I would estimate each kid deserves based on how much science they seem to have learned and my impressions of their general work quality. So I feel fine about putting these grades on the report card, and matching them up with the rubric according to my idea of what the rubric means. All these philosophical-mathematical questions just give me the opportunity to fine tune my grading for the future, and to really think about what my grading is communicating to kids and saying about my expectations.
In the end, that's what I did. My colleagues used their own rubric-percent conversion chart. Does this help parents or high schools make sense of our report cards? Nope. But they'll just have to deal with that until we all sit down and hash out a final, FINAL system.
As for my own grading, I've decided that I need a new homework system. This fall, I used the 1-4 rubric because it allowed for very fast grading, and took into account multiple factors (as described above). I get about 100 sheets of paper almost every day, so even if each gets just 2 minutes of attention, that's more than 3 hours of grading! Woe is me! The solution to this might be to collect & grade homework less frequently, and spend a little more time on each assignment. I could develop a meaningful percentage grading system if I only graded 200 papers per week; the other days I could just check off whether or not the child did the assignment and go over it in class. Problem with that is that reading homework is a way for me to get information about the kids' misconceptions, and reading it only twice per week would limit the information I get. Another option would be to collect every day but grade only a selection of the questions... but I don't think that would really save me much time compared to collecting every day and grading all questions.