Feature request: Don't pair #5
Comments
|
That you can actually do with a distribute rule. Make a column and give each pair (or larger group) you don't want together a unique key (number, letter, word, ...) and assign a distribute rule to the column. That will then cause all the students from, for example reviewer 1 to be assigned to different groups (consider making the rule a high priority one, it shouldn't interfere with other rules much, but other rules could interfere with it) |
|
That is indeed what I did. However, it was cumbersome. I needed seven columns, named avoid1 to avoid7, to fully specify what I had previously specified in just two columns, and it's hard for me to estimate how much time it took me to manually translate the two columns into the seven columns because it took me multiple sessions. |
|
I may be misunderstanding something in what you are doing then. Do you have
seven distinct groups they have been paired with? How would you like to be
able to specify this information (what is the natural way to record the
information)?
…On Tue, Oct 2, 2018 at 8:20 AM vivarose ***@***.***> wrote:
That is indeed what I did. However, it was cumbersome. I needed seven
columns, named avoid1 to avoid7, to fully specify what I had previously
specified in just two columns, and it's hard for me to estimate how much
time it took me to manually translate the two columns into the seven
columns because it took me multiple sessions.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#5 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AArz9EpvPzREl8G3uiQ3O4dqrMCUh5VXks5ug4QggaJpZM4XCuqb>
.
|
|
No, I have an interlocking web of students who must avoid each other. Student A was an anonymous reviewer for students B and C, so needs to avoid students B and C. In the end, pairing B and C is good. Neither one has peer reviewed the other. So I can't say that A, B, C are one distribute group, or else I miss that possible pairing. I have to make A and B a distribute group and then I need a new column so I can make A and C a distribute group. |
|
Ok, yeah, that looks like it would need a new rule. I'll think about how to
to that, if you can provide an example file with the info in a form you
would like to provide it in, that would be helpful
…On Tue, Oct 2, 2018, 8:36 AM vivarose ***@***.***> wrote:
No, I have an interlocking web of students who must avoid each other.
Student A was an anonymous reviewer for students B and C, so needs to
avoid students B and C.
Student B was an anonymous reviewer for students D and E, so needs to
avoid them.
Student C was an anonymous reviewer for students A, E, and F (three
students because A and F wrote a report together), so needs to avoid them.
In the end, pairing B and C is good. Neither one has peer reviewed the
other. So I can't say that A, B, C are one distribute group, or else I miss
that possible pairing. I have to make A and B a distribute group and then I
need a new column so I can make A and C a distribute group.
—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
<#5 (comment)>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AArz9M_dntgglbj-jnFH9eP1KREAb6UJks5ug4gPgaJpZM4XCuqb>
.
|
|
(I sent Tom the files by email.) |
I would like a feature where I can name people not to pair with a particular person. In my use case, I have students working in pairs, and I have also assigned anonymous peer reviewers. I would like to assign the students to new partners, but I don't want them to be partnered with someone who has been their peer reviewer.
The text was updated successfully, but these errors were encountered: