De-Grading, part 4: Curving Grades

“Curving” grades is supposed to be a no-no, but many of us have ways to adjust grades, particularly on tests. Because teachers have many things to do, we often choose the simplest method of adjusting grades, even though (if we had time to think about it) it isn’t the fairest or most logical method.

Here are some examples:

Watch for Hidden Curves

Scenario: I give a test to my class. When I start grading, I realize that the test was much harder than I intended. The class average works out to 48%, the high score 88%.

I don’t want this one test to skew their grades for the quarter. Some teachers would call the high score an A and raise the rest by the same number of points. Not exactly a curve, but it’s a common way to adjust scores. To make a true curve, I’d have to make 48 a C and spread the rest of the scores above and below with most of the scores in the middle range.

But I’m opposed to curving grades, so I decide that instead I’ll make the next test much easier, and the scores will all even out. That way, nobody’s grade will be hurt by the one ‘bad’ test. The next test is much easier; the average is 88%, the low score 74.

But my decision has the opposite effect from my intention. The ‘bad’ test ends up being the determining factor for every student’s grade. Using a curve would have been fairer.

Ninety percent of what?

Why is 90% an A, anyway? Where does the idea that a certain percentage is ‘passing’ come from, while one percentage point lower is ‘failing’? Does this reflect some innate organization in the human mind? Obviously not.

Grading practices in other countries vary, but few use a rigid percentage scale. American teachers are obsessed with percentages, though. For one thing, they seem more objective. The other reason is that grading by percentages requires less thought. Teachers don’t have a lot of time for that, and even if they did, many districts (like mine) mandate a percentage scale.

But percentages are subjective . Excluding standardized tests, what I put on my tests and quizzes is what I decide to assess. Other teachers may make very different decisions.

For example, on the foreign language hallway, another teacher and I give weekly vocabulary quizzes, twenty words. She assigns twenty new words on Monday and quizzes students on Wednesday.

I also assign twenty new words a week, but only half of them are on my Thursday quiz; the other ten words are older words. I like to recycle old vocabulary to make sure they haven’t forgotten them.

The average score on her quiz is usually 30-40%. On mine, it’s about 65-70%. She complains constantly that her students don’t memorize.

But it’s normal for us to forget about half of new material we’re trying to master; it takes a lot of repetitions to get new words in our long-term memories — at least six, and possibly more than twenty, or even (as one of my students says) ‘a billion.’ That’s just the way our brains work; it’s much easier to forget than it is to remember.

This is just  my opinion, but if 60% represents the lowest passing grade, 60% of your test should be material that the lowest performing student knows. The newest material should comprise no more than 20% of the test; those questions are what separate the C students from the B’s and A’s.

But why spend time grading answers that everyone got right? Why test easy material that every student knows? To boost their egos? To boost my own ego? To make other teachers think I have a secret method for teaching vocabulary?

If my neighbor could accept the notion that curving grades is fair and sensible, she might look at her results and decide that 40% is a C. She could examine how her scores are spread, and select a cut score for failing.

Instead, she compensates by giving her students many points just for completing assignments, for handing in pieces of paper. Averaged in with their low quiz scores, this evens out her grades and makes her students look successful. And it teaches her students that it doesn’t matter if they fail quizzes; pieces of paper handed in on time are what really count.

I don’t fault her. She’s under the same scrutiny as the rest of us: if she fails too many students, her principal will decide she is being unfair, and possibly change her grades. This is not a hypothetical scenario; it happens all the time.

Teachers are Professionals

If you have been teaching for several years, you have most likely developed some consistency in your tests and quizzes. You know what your students are capable of, and your tests don’t swing from too easy to too hard. As you teach your lessons, you do informal assessments of how well your students are mastering the material. You can see what you need to re-teach, and what needs less time.

From this perspective, you have the experience and professional discretion to write tests that measure what your students should learn, and to grade them as you see fit. You understand that a test is just as much an assessment of your teaching as it is of your students’ progress. You know when they are slacking — and when you are.

You are a professional. Design your own rubric, or scale, or method of assessing your real-life students, rather than trying to fit them into a fixed scale designed for hypothetical students taking imaginary tests created by fictional teachers.

Next: Gordian Knot

Advertisements

1 Comment (+add yours?)

  1. Trackback: De-Grading, Part 3: Zero « edaxicon

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Enter your email address to subscribe to this blog and receive notifications of new posts by email.

Join 14 other followers

%d bloggers like this: