Yay! And nice to see that they actually listened to the students and didn’t wait for the parents to get involved.
@sylvan8798 I was unclear, perhaps.
I specifically said “based on the grades of the rest of the class” because that, to me, specifically excludes a curve where everyone gets a boost because the exam was too difficult, or where the intention is that everyone be able to successfully do, say, half of the problems correctly, to gauge the upper limit of the students’ understanding.
Glad the OP had a happy resolution!
^I guess I still don’t get it. Once the instructor decides to put a curve on the grades, then the curve is usually based on the grades in the class. IOW, if one applies a factor to the grades, it is probably determined by the highest grade and applied uniformly to everyone else. When I was an undergrad, I was frequently known as a curve buster because I would ace the exams and leave no room for a curve.
Is the term “bell curve” what I should have used, one in which the grades are forced along a distribution of x# of As, x# of Bs, etc?
^Yes, that would be a distinct type of curve, which, I agree, is not necessarily the ideal way to go about it. In general , there is no particular reason why there should be x# of As, etc. in a particular class.
There was another thread about curving recently. I’d like to ask how many of the people who strenuously object to any kind of curving have ever written exams or looked at the distributions of class scores themselves?
If the college board has to employ statisticians and pre-test their rather simplistic questions on many thousands of students in order to get them to be the right difficulty, how is a teacher or professor working alone with no opportunity to pre-test the questions going to be able to precisely match the difficulty of exams they made up from year to year?
Also, what is meant by “forcing” grades into a distribution? Is making an 89 a B and a 90 an A not forcing a distribution? Lines have to be drawn. Often there are clusters and gaps which make it obvious where lines should be drawn. That makes more sense to me than setting arbitrary cut-offs which give students whose performance was essentially the same different grades.
Teachers should be looking at the mean, not an outlier, if they are going to curve.
I agree.
I suppose they’re distributed already, or can be. But I take issue with failing (or passing) students based on their grades relative to each other rather than to their mastery of the material and application of concepts, or whatever the expected bar for passing is.
We had programs that could read two programming assignments and come up with all kinds of metrics like numbers of variables, functions, lines of code within functions, etc. It was pretty decent, flagging obvious cheats. I was a Graduate Teaching Assistant 30+ years ago and we used this on a mainframe (for PL/1 code lolz).
Right before this, programming assignments generally counted 40-50% of the course grade, meaning you could really nuke the tests and still get a B. Then there was a wave or two of cheating (suffice to say that the XYZ Student Association (where XYZ denotes a specific country that had several hundred students on campus at the time, most in CS, had a magnetic tape archive of every assignment ever given in the last 20 years…) then assignments were dumbed down and only counted for 20% at best.
This was counter productive, because in my days of 40%-50% assignments, you really learned how to write code quickly and correctly. We might have been a directional state school at the time but had a very good CS program with challenging assignments. I still write code 30+ years later. So there were some mini-student revolts and program count percentage was restored… Eventually ‘software engineering’ took over and group projects became more common. This has always been a CS issue, down to the point of having intro CS classes do assignments IN CLASS (the horrors) to avoid cheating…
^It’s not easy to see how that transfers to other types of courses.
I notice that posters here blaming the teacher haven’t suggested methods they would consider more effective…
Consistency across the department - whichever way, absolute or curved up - would be more effective. Apparently the HS agreed.
In colleges? Should all the instructors teaching the same course be using the same lectures, exams, labs, grading system, etc.?
This isn’t college, so why do people keep talking about college?
I was talking about HS.
Our school almost never curves but occasionally will give back a few points if students do test corrections. This makes a lot more sense to me because that way the student has to go back and work at what they missed until they can get the correct answer rather than just being given a few points for doing a little better than their peers. My daughters AP Physics C had a recent test where 70 was the highest grade but it was incredibly difficult material taught in a short period of time. With test corrections she can get to a low B but will still maintain an A for year with high lab grades and the other tests and quizzes. I have always talked to my kids when they get saved with a high homework grade or other projects and such that they should not expect that in college and that they have to give it all for the tests in those classes that don’t have other grades to pad the average.
The NYC DOE recently updated its academic polic, which it sent out to all principals to trickle down to all staff:
Review Updated and New Academic Policy Guidance Documents
All schools
In preparation for the 2016–17 school year, you should review the following new and updated resources on the Academic Policy Intranet pagehttp://intranet.nycboe.net/Accountability/APR/, as applicable:
New Academic Policy Guidance Documents:
- Grading Policy (all schools): The NYCDOE has created a Grading Policy Checklist and FAQhttp://intranet.nycboe.net/NR/rdonlyres/B3FC1669-E14D-473D-8AE5-654621456291/0/AcpolicyGradingPolicyChecklist.pdf to guide you as you develop and improve your school’s grading policies. In your grading policies you must outline when and how grades are awarded, including, at a minimum, how mastery is measured, how courses are graded, the scale of marks awarded as well as a timeline of when students receive their grades.
Please note that you have discretion in deciding which specific measures are factored into students’ grades. However, determinations of pass or fail must be primarily based on how well your students master the subject matter, concepts, content and skills addressed in a class or course. Students cannot pass or fail primarily based on non-mastery measures.
Documenting Course Content (all schools): As you plan your grading policies for the 2016–17 school year, you should review this one-pagerhttp://intranet.nycboe.net/NR/rdonlyres/C38DDF59-EF90-4035-A1EF-9036198C899E/0/AcpolicyDocumentingCourseContent.pdf, which provides information and best practices for storing your schools’ course records from year-to-year.
IF you work for the NYC DOE, you may need to log in with your DOE email account to download the PDF.
For the 2016-2017, teachers cannot state in their grading policy that the course is graded on a curve or set quotas
the policy now states:
Grading on a curve does not imply that there is a quota percentage of students receiving each particular grade. This example is conflating two distinct grading practices.
Will be interesting to see the outrage when teachers decide this policy means they aren’t allowed to curve, and AP students are being failed left and right because they only scored 50 or 60 percent on AP questions.
Consistency is always better than inconsistency as far as education. My spouse always coordinates with her colleagues with courses that she designed. They teach the same material as designed for the course and use the same material across sections and instructors. One person creates the exam used by all.
Gosh what a radical idea.
I don’t see it happening. A curve there (adding 25 points to all or whatever) isn’t the same as having a class where the grade distribution is forced into certain grades.
Can you give any example of a class where the grade distribution is forced into certain grades? How exactly do you think this is being done? Some posters keep talking about this, but I’ve never seen it happen and I’m not sure I can envision a case where, say, a professor thinks, Oh, I need to fail 3 students to make quota of Fs, well this guy scored 40 points higher than the other two I’m failing and only 2 points lower than someone I’m passing, but I’ll just fail him because I need to fail 10% of my students. Have you seen cases like this? (Of course, that scenario could very well happen with “absolute” standards of mastery passed down on tablets from God).
I’m wondering what the definition of mastery is and where is is coming from. On a thread a few months ago I was roundly criticized for expressing surprise that only perhaps 10 of the 100,000 students who take AP calculus exams each year manage to solve every problem correctly. So, do we flunk all but the 10 who can do all the problems, because the others clearly haven’t mastered the material–they couldn’t do all the problems posed, which they were supposed to have learned how to solve. And these problems are usually pretty straightforward compared to more advanced college coursework. And if not, where do we draw the line of what “mastery” means? And does it make more sense to draw an arbitrary line at arbitrary percentages or does it make more sense to award grades according to which students perform the closest to one another?