Tip:
Highlight text to annotate it
X
(male speaker). And, although on the other end
of the scale it might not be as clear cut.
That's the case, it may be that in fact students you know,
that the knowledge was there and available and they were able
to learn it and achieve it.
So, I don't think it's as clear cut that that you should curve
both ways philosophoically even.
(Marilla). Well, you were doing fine
until you hit that philosophical thing.
There are pragmatic reasons for curving the grades, and there
are philosophical reasons for curving the grades.
And okay, caveat emptor, I don't believe in curving the grades.
I believe that, and you stated it very nicely that if you have
set your expectations and you set and you've written a test
that you believe is an appropriate assessment of what
you expect from the students and they don't do well on it, rather
than curving the grades I go back and I reteach the material.
And give them an opportunity to come up to my expectations,
rather than to lower my grades into theirs.
And I did this once with one of my classes you know,
I write great tests.
That was funny.
I don't, I mean I do, but I write okay.
So I gave this test and I always have this essay question
on the test.
And, ugh, the students just bombed it.
And I thought you know, I cannot let this pass.
So I said, I came in to give the tests back and I said okay,
the performance on the last question which was the
essay question was not what I was hoping for.
And this is an important point, so I'm going to let you take
your tests back and rewrite the last question.
And turn it in to me, and I will give you back half of the points
that you missed on that.
Okay, well first of all there was this huh gasp, and then
you know, I said you don't have to do this.
But, you know I think that I want, I would like you to do it.
They took it home, they rewrote, they brought it in the next
class period and they handed it in.
And this is kind of the hokey part, but it was this really
made an impression on me.
One of the students handed her test back to me and she said
you really do care whether we learn this or not don't you?
And I was like yes I do.
And you know fade to black.
But it was you know, it was very touching that they really
appreciated the fact that you know, I wasn't going to just
like let it slide.
So I would say philosophically, I don't believe
in grading, curving.
I don't believe in grading, well...
I don't believe in curving because of the philosophy
that's behind it.
I understand the pragmatics that some people have when you have
an instrument that's not a good measurement instrument.
But that's the fault of the instrument, and I think
you should need, you need to work on the instrument
rather than curving the grades.
(male speaker). [unclear audio].
(Marilla). Yeah, I don't think curving
the grades is the way to deal with it but certainly
I understand what you're saying.
(male speaker). From a practical point of view
I don't do it on the individual exams.
If there is one low one, certainly you have the points
to bring it up, but if you have a high one there is just no way
they'd allow you to do it.
So at the end of the semester I'll do it, and fortunately
or unfortunately [unclear audio] and that scale is 75%,
so that's something I can say if we want to go up,
this is kind of what the standard is.
(Marilla). Right, actually that's
an interesting, you raised a really good point and that is
that sometimes the standards are set for you by an outside
source and therefore your, I believe our responsibility
as instructors is to help our students achieve
those standards.
And what I want is all my students to be able to pass
or to do that.
And then which case, grading on a curve is not a good thing in
my class because you want them to come up to a set standard
and so that then, that's criterion reference grading.
So that's a really good point.
Oh, sure.
(male speaker). Also, it occurred to me,
I'm concerned about extra credit.
It occurred to me that's one way we could deal with unexpected
answers, but extra credit in general whether that's a way
to deal with pain, or that [unclear audio].
(Marilla). There may be other,
but lets talk about, well we'll talk about,
we'll talk about that as a group.
Does this group, didn't give a chance to give
their second one, so do you have another one you want
to add up there?
(female speaker). I think we got them all.
(Marilla). You got them all?
Alright, okay so we have student's pain I think centers
around this idea of fairness and expectations.
And so if they are evaluated in a way that they
can recognize--well I'll tell you I'll pick out that part
of it--and they see is uniformally applied,
that does a lot to mitigate their pain.
So let's talk about that.
Okay, let me see if I can get to that.
Alright, we're going to do, let's see here.
Okay, this will deal with this issue.
Here's a little mini-course in creating fair
and consistent grading.
So this I think is really important.
And so this is like straight out of measurement theory.
I used to call these the four R's of assessment.
But nobody appreciated it so now we get the four principles
of assessment.
Okay, so the first principle, these are the four.
These are the four, sound, measurement practices.
You need to have a measurement that's valid, and I'm going to
explain each one of these in more detail, reliable,
recognizable, and realistic.
See how three of them are R's?
The first one I wanted to call relevant, but you know,
then the measurement people got on my case and said no
it's not relevant, it's valid.
Okay, alright so what do they mean?
Okay, valid is that the assessment measures what
it says it measures.
it says it measures.
Let's see here.
No, this one doesn't have one of those.
Okay, well, we'll just go on.
Reliable means that the measurement is consistant.
I'm going to explain each of these in more detail,
and show you how you do that in your classes.
So reliable is that the measure is consistent,
measurement is consistent.
Recognizable is that the assessment is familiar.
And realistic is that the amount of assessment
done is reasonable.
Alright, so if you can create measurements or tests or essays
or activities that meet these four standards, then you are
going a long way toward the fairness and consistency issue.
So what do they mean?
Okay, so here is valid, an instrument is valid if
it's based on the learning objectives.
It measures what you say it's measuring, and one way
to do that is to take your learning objectives and base
the assessment on that objective.
So, I'll give you the worst example of a non-valid test
I have ever heard of.
And I tell you this is the true, this is a true story.
It was in a graduate class, and the instructor had a long
reading list of articles, technical articles you know,
in the journals.
And on the test, what the students had to do was match
the title of the piece with the author.
This is a classic example of a non-valid instrument.
You can imagine the reaction of the students, okay.