Four self and peer assignments, each using adaptive release rules.

On the CHEP Community Team Dr Poshak Gandhi asked some questions about peer assessment:

“My module has 170 students, split into 4 groups of about 42 students each.


I want to create an assignment for peer assessment, with each group having a separate submission deadline.  I don’t see the group-wise submission deadline option on the peer assessment page (see attached screenshot).


How do I implement this? Am I missing something? Thanks.”

Poshak Gandhi: Dear all in General, Question on Groups and Peer Assessments in B… 
posted in CHEP Community / General at Oct 5, 2020 12:59 PM

So, I have spent some time looking into this and here are the results of my testing.


Too long to read? Skip to the conclusion.


The plan

My plan was to set up four groups, and four self and peer assessments, then use adaptive release to make each assessment only available to each group. While I was fairly confident this would work I needed to run through a whole end to end test to be sure. To save a little time, I used the group set feature to randomly enrol the 28 test student accounts across four groups.

Links to guides that show you how to do this are in the paragraph above.

The setup

In these screenshots you can see my groups, assignments, and an example adaptive release rule:

Testing

Part 1, students submit their answers

The first part of testing was to submit an answer on behalf of each of the 28 test students. To make it simple on my side, the answer I gave was the username I was using and what group I was in. I wanted to be sure that I would not find that students had to mark submissions of other students who were not in the same group.

I left one test student without submitting, to see what would happen.

Submissions to group 1’s assessment.

In the above screenshot, I am viewing as an instructor the submissions for the group 1 assignment.

Since adaptive release only affects to whom the assignment is displayed, the assignment itself does not know who should be taking part. For this reason, we see a lot of students who have not started it. This is because due to the adaptive release rule they cannot see it.

Lesson 1: Using this approach will mean that instructors will see non-submissions from students not in the group. It won’t be easy to identify students who have not taken part when they should have done until the assessment is complete.

Part 2 – Students evaluate the submissions of their peers

The next stage was for students to evaluate the submissions of their peers based on the criteria I set. To make it simple for testing purpose I set it so that each student evaluated one other student against a single criterion. I also disabled self-assessment so that students would not be assessing themselves.

An example screenshot of a student completing an evaluation.
Screenshot of a test student completing an evaluation.


As I went through the process I noted which student was assessed by each student, I wanted to be sure that a student would not find themselves being asked to evaluate a student from a different group.

This table summarises my results.

Select to reveal the table. table, th, td { border: 1px solid black; border-collapse: collapse; }
Group Account Submitted? Who they reviewed
Group 3 btrain1 x btrain17 Group 3
Group 3 btrain10 x btrain16 Group 3
Group 3 btrain11 x btrain22 Group 3
Group 4 btrain12 x btrain25 Group 4
Group 4 btrain13 x btrain12 Group 4
Group 4 btrain14 x btrain24 Group 4
Group 2 btrain15 x btrain20 Group 2
Group 3 btrain16 x btrain11 Group 3
Group 3 btrain17 x btrain10 Group 3
Group 4 btrain18 x btrain14 Group4
Group 2 btrain20 x btrain26 Group 2
Group 2 btrain21 x btrain32 Group 2
Group 3 btrain22 x btrain33 Group 3
Group 2 btrain23 x btrain15 Group 2
Group 4 btrain24 x btrain18 Group 4
Group 4 btrain25 x btrain 13 Group 4
Group 2 btrain26 x btrain9 Group 2
Group 1 btrain27 x btrain28 Group1
Group 1 btrain28 x btrain35 Group1
Group 1 btrain29 x btrain27 Group 1
Group 1 btrain30 x btrain 8 Group 1
Group 4 btrain31 Did not submit Nothing!  Note that the student is not told they don’t see a submission to evaluate because they have not submitted originally.
Group 2 btrain32 x btrain23 Group 2
Group 3 btrain33 x btrain 1 Group 3
Group 1 btrain34 x btrain29 Group1
Group 1 btrain35 x btrain31 Group 1
Group 1 btrain8 x btrain34 Group 1
Group 2 btrain9 x btrain21 Group 2

Lesson 2: With this arrangement, all students who made a submission were able to evaluate someone else in their own group.

Lesson 3: Students who do not submit cannot evaluate a peer. This is not made clear to them so should be kept in mind when explaining the exercise to students.

Part 3 – The instructor checks evaluations and sends results to the Blackboard grade centre.

The final part of a self and peer assessment begins once the evaluation period has ended. At this point, the instructor can review the answers and evaluations, and then choose to send the peer evaluation results to the grade centre in Blackboard.

Since a large number of students are shown who would not have been in the group that the assessment was meant for, I found it useful to organise by completed evaluations as shown in this screenshot.

Screenshot showing the results of a self and peer assessment.
Screenshot showing the results of a self and peer assessment, organised by the Evaluated column.

Lesson 4: Instructors should continue to be aware that all students are shown in each assessment, even though they would not have been able to access it, this again is because adaptive release only affects the assessment visibility, groups are not considered by the assessment itself.

Part 4 – Viewing results in the grade centre

Grade centre screenshot
Viewing the full grade centre, note that “patchwork” of results due to students only taking one of the assessments.

When we view the full Grade Centre we see all the assessments and all the students. A student will only have one score for one of the four assessments.

To make this easier the instructor can create Grade Centre “Smart Views” based on the groups she has created in the course. Furthermore, these Smart Views can be favourited so that shortcuts to them appear in the Control Panel.

Screenshot of the Grade Centre in the Control Panel, with favourited smart views.
Screenshot of the Grade Centre in the Control Panel, with favourited smart views.

When the instructor uses the Smart Views, reviewing the results is more manageable.

Grade Centre screenshot, a smartview for group 1 has been selected.
Grade Centre screenshot, a smart view for group 1 has been selected. This means only students in Group 1 appear. The only have scores for assessment 1.

Lesson 5: Use Smart Views to make viewing the Grade Centre more manageable.

Part 5 – Viewing results as a student

Students will have two main ways to view their results. First, if they enter the assessment once the evaluation has completed they see their answer and their evaluation:

Screenshot showing what a student see when checking into the assessment after evaluation is complete.
Screenshot showing what a student sees when checking into the assessment after the evaluation is complete. Note that the feedback to the learner is me typing random letters.

Alternately students can check their Blackboard scores, to see all marks from the course. I was surprised and happy to see that only completed scores are shown, so the students should not be distracted by assessments they could not complete:

Screenshot showing a students marks on a course.
Screenshot showing a students marks on a course.

Conclusion

Using Blackboard groups and adaptive release is a practical way to make separate self and peer assignments appear only to certain groups of students. Creating group smart views and favouriting them makes using the Grade Centre more efficient because this allows the instructor to focus on one group at a time. If different instructors were reviewing different groups this will be extra useful.

Here is a summary of the lessons:

Lesson 1: Using this approach will mean that instructors will see non-submissions from students not in the group when reviewing submissions. It won’t be easy to identify students who have not taken part when they should have done until the assessment is complete. This is not a show stopper but should be kept in mind.

Lesson 2: With this arrangement, all students who made a submission were able to evaluate someone else in their own group.

Lesson 3: Students who do not submit an assessment cannot then evaluate a peer. This is not made clear to them so should be kept in mind when explaining the exercise to students. Clear communication is vital with this tool since the feature itself does communicate particularly clearly the expectations to the students.

Lesson 4: Instructors should continue to be aware that all students are shown in each assessment, even though they would not have been able to access it, this again is because adaptive release only affects the assessment visibility, groups are not considered by the assessment itself.

Lesson 5: Use Smart Views to make viewing the Grade Centre more manageable.

Final thought

The self and peer assessment tool takes a little practice to become confident with. A major consideration is that the system does not tell students when the submission stage ends and the evaluation stage begins, so careful oversight and communication is essential. Therefore if you plan to use it I very strongly recommend running a very short formative exercise with it first so that you and your students will be confident with it.

Using the Blackboard ‘Self and Peer Assessment’ tool with groups

Post navigation


What do you think? Leave us a comment to share your thoughts...