# Grading: evaluative or coercive?


**Introduction**

I have had many opportunities to think about grading and assessment this
term. I am currently running a course called [Introduction to
Computer-Supported Collaborative
Learning](http://bit.ly/intro-cscl), and have thought a lot about
incentive mechanisms and [badges](http://bit.ly/badgepaper4) with
[Monica
Resendes](http://test.kmdi.utoronto.ca/graduate/students/#resendes), who
is co-facilitating the course with me.

I am also preparing to design and teach a course in ICT for development
at Scarborough Campus in the fall, and as part of my preparations, I
attended a two-day course design institute for graduate students with
the [University of Toronto Center for Teaching Support &
Innovation](http://www.teaching.utoronto.ca/about_ctsi/servicesexpertise/ctsi-workshops/workshops.htm#Course%20Design%20Institute)
(CTSI), led by [Megan
Burnett](http://www.teaching.utoronto.ca/about_ctsi/contact/assistantdirector.htm).
This institute was a great experience, giving me a lot of new ideas, and
new perspectives on things.

During the same time, I was enrolled in an online class on
Computer-Mediated Communications, and I brought many of the ideas from
the CTSI institute into the class discussions, including this discussion
of grading.

**Evaluative grading**

According to Megan, when you design a course, you should start with
identifying your specific course goals and objectives, which can be
content-specific or skill-specific. For example, an undergrad class
might aim to both give you the ability to identify the main historical
figures from the 20th century, and the skill to use proper MLA citation
(as well as many other goals, of course).

Only when you have these goals, do you look at assessment, and ask: what
kinds of assessments are appropriate to test if the student has acquired
the content knowledge and skills that you hoped. The key point is that
you should only test this, not all kinds of other things - ie. if your
goal is to teach your students to write proper English, then you should
correct the English in their essays, not otherwise.

**Coercive grading**

However, thinking about the grading structure of many courses I have
taken, I realized that there is a very different way of using grades -
not to evaluate, but to compel behaviour.

You could argue that all grades are too some extents designed to be
“compelling” - humans are not like inert objects that can simply be
measured, without it leaving any effect - clearly one of the functions
of giving grades is to compel us to work harder (although in some cases,
it can have the opposite effect). However, there are situations in which
the grading has no evaluative aspect at all, and it functions merely as
a carrot and stick to get students to do something - typically something
the teacher believes is helpful to their learning.

If, for example, you believe that students who write at least two posts
per week learn more than students who don’t, and you allot 15% of the
grade to “participation” - ie. writing at least two posts per week, then
you are not really “assessing” students ability to write two posts per
week, you are simply using the grade coercively to “punish” those who do
not do this.

Perhaps there are students who differ from the average, and actually
would learn much better by not participating in the forum at all? In a
traditional undergraduate class, you can encourage them to come to
class, but it is their choice, and if they choose not to, and do
excellently on the essay and final exam, you’ve got two choices: admit
that your assessment methods do not properly capture the learning you
wish to assess, or be very happy about their success.

(Another issue is of course that since there is no pre-test, it is
difficult to assess the "value added” of your specific course - perhaps
they had this knowledge and skills before they joined your course. But
at least your final assessment should make you confident that they have
reached a level of knowledge and skill that you desire).

**Online courses**

****What’s interesting is that this “coercive grading” tends to be much
more prevalent in online courses than in offline courses, partly perhaps
because it is easier to do (with the wealth of statistics being
captured, and made available to the teacher), and partly because the
teacher is more afraid of student disengagement.

At the CTSI workshop, they told me that UofT policy strongly discouraged
including class attendance as one factor in grading. Yet almost all
online courses include “posting every week” as one of their explicit or
unspoken factors of assessment.

When I brought this into the discussion in the online course, I received
some very interesting feedback, implying that people felt very
“defensive” about the term coercive grading, which seemed to have very
negative connotations. For me, this concept of evaluative vs “coercive”
or “compelling” grading is a really useful thought tool, and does not in
itself imply any value judgment.

In thinking about it, the first question should be whether this is a
valid distinction, or whether there are other categories that should
also be included. Another question altogether is whether, or in which
circumstances, coercive grading might support the learning process.
However, this is a discussion that we could not even begin to have,
without having clarified the terms first.

**P2PU**

This is also a very interesting discussion when thinking about
[P2PU](http://p2pu.org/), which has no coercive grading. We also have no
evaluative grading right now, although we are developing an
infrastructure for evaluation and recognition of learning achievements,
focused much more on competency-based assessment. This really changes
how we can implement different pedagogies, and the amount of leverage
that a course organizer has over participants.

Talking to a professor at UofT about his or her course, the discussion
is structured very much in terms of power: “Then we’ll have them do
this, and then they’ll have to do this”. The students have paid good
money for the course, and need a decent grade, so they have no choice
but to comply. Whereas at P2PU, it is free to attend, there is not much
external value (currently) in completing a course, and often you don’t
know the people you are studying with, or the course organizer, so there
is less of a social commitment.

The unrealistic ideal might be that students would be focused on their
own learning goals (which they would have clearly outlined before the
course), and would, with the support of the course organizer and other
participants, pursue the learning that would be most useful for them to
reach their learning goals, while at the same time supporting the
learning community in the group. Of course, we know that this often does
not happen, and sometimes we might wish we had the power to “compel”
certain behaviour… But I think an environment without that power is a
great place to experiment.