Community

cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
sinman
New Member

Seeking Course Analytics Advice

Good evening all,

I have an inquiry about course analytics. Our faculty really want to utilize the tool; however, a few do not see the usefulness of it. We had a faculty member create many groups with discussion assignments within them. However, it seems that on the course analytics page, it isn't possible to see student activity such as commenting in a discussion forum (particularly if it is within the student group).

Also, a functionality question: If the professor has a discussion board published as an assignment, but it is actually migrated over from an old course (so the deadline is in the past), will that reflect negatively on the student regarding the analytics?

Thanks to all for any advice on how to utilize course analytics optimally!

Sarah

Tags (2)
10 Replies
Stef_retired
Community Team
Community Team

Interesting discussion prompt,  @sinman ​! I've shared this with the Higher Education​ and K-12​ groups to enhance its visibility among like-minded Canvas users.

d_ellis
Community Contributor

Hi Sarah,

I have found the course analytics page not especially useful for staff; it is often too broad an overview to drive changes to teaching or course design.

What are you looking to see in the course analytics page? As far as I can tell, student posts to discussions within groups do get counted and recorded as "participations" in the course analytics page.

orwinr
Community Champion

Hi  @d_ellis 

You are correct about group discussions and posts not counted as participations. We have had an ongoing issue with the way groups work when doing group discussions. I kind of think of groups in the same way I think about Las Vegas, "What happens in Vegas stays in Vegas." Smiley Wink Groups function in the same way, "What happens in groups stay in groups." Smiley Sad I have had students and faculty members complain for years about notifications from group discussions not being sent because the discussion is in a group, faculty members in particular. To me, groups leave a lot to be desired at this pointSmiley Sad

d_ellis
Community Contributor

? I thought I said the opposite, that activity in groups (announcements, new discussions or replies, etc.) does count in the course analytics report - that is, it will contribute to their "participations" and "page views" counts on the main site. Group activity also shows up in a student's access report. In the particular case of course analytics it doesn't appear that what happens in Vegas stays there :smileyshocked:  Sorry if I misunderstood your reply.

I haven't tried out the notification behaviour within groups though - it sounds like a pain Smiley Sad

orwinr
Community Champion

Sorry Damon. I misread your post and obviously haven't tested the behavior of group discussions in analytics for a while. I will have to test again and see if the analytics and notification behavior has changed since I last tested. I will post back up here and let you know what I find if you are interested.

Good morning!

I'm looking to see how many posts a student has made -- both to a course discussion as well as a group discussion. It seems that we currently cannot do that aside from manually counting each participation. bleh

Sarah

Stef_retired
Community Team
Community Team

 @sinman , have you seen the solution  @James ​ created to tally discussion participation across an entire course? You'll find it at How to Count Student Discussion Posts​. The caveat here is that it will only pull in posts made to course discussions, not to group discussions--James says it's possible, so if someone wants to develop that supplemental resource, we'd love to see it.

James
Community Champion

At the time I wrote it, I had never used group discussions. I have now and they are handled differently than course discussions, so it will probably be a little bit more involved than just changing the API call. I haven't looked into it as far as modifying the discussion counts spreadsheet, but it looks like it will add another level of nesting to the process. Each group discussion actually consists of a separate discussion for each group. So, when a group discussion is encountered, you would need to fetch a list of the group discussions that are associated with it and then iterate over each of those. They would all get combined into a single group discussion in the spreadsheet.

Canvas is similarly clunky when it comes to group discussions. If I remember correctly, when I go to the discussions page, it shows 0 unread messages for group discussions, but if I click on the discussion, then I can see how many are unread in each of the individual group discussions.

Hi James!

This is amazing - the spreadsheet you created! But I do see what you mean about it not grabbing posts made in group discussions.

The course and group discussions seem to be located in different places. While you can see participation counts on the access report for individual students, you can't actually see how many posts a student contributed. You do bring up a great point in your post though. Most faculty will not be able to do this so we will need to generate the report for them as needed. But heck, your instructions made it so simple. If I can do it, anyone can. Smiley Happy

Thank you!

Sarah

James
Community Champion

I've got another spreadsheet that I've been using myself this semester rather than the one that does it for the entire semester at a time. It doesn't do groups either, but for every course discussion, I've gone through and documented first post date, last post date, days active, number of posting sessions, number of posts, total words, total characters, likes received, number of replies to the post, and then -- if you have admin/masquerading capability like I do, it adds the number of likes given, the number (and percent) of posts read/unread -- and does this for each student, but only for one discussion at a time.  It also looks at each posts and breaks down by similar characteristics, although I don't use it when determining grades.

Then when the student comes and says "Why did I get a 'Poor' (45%) on this?', I can say you waited until 2 hours before it was due, made your post, made one reply, didn't read 97% of the responses from other students, and spent less than 5 minutes on the week-long discussion.

I grade them holistically and go through and rank students based on certain features and that generally narrows the grade down to 1 or 2 possibilities. Then I go in and read them in SpeedGrader before making the final decision.

I've collected those all semester long, except for the ones that are group discussions, but those were for working on their projects.

The last week's discussion was a course reflection. Overall, the comments were positive, but the number one thing they disliked were the discussions -- claiming it was too hard to get an "Awesome" (105%) [although 14% of the posts were graded that way] and they had no idea what was expected out of them and other classes just made them post an initial response and two replies - it didn't matter what they wrote. The day of the final, I pointed out that I had given them a handout the first day of class with what I was looking for in the discussions and explaining the grading system. The biggest problem is that student's either don't read, listen, or retain the information you give them (this is a general education statistics class taken mostly by people who hate math but have to have it to graduate). Although many didn't care for the discussions as a whole, those who commented generally did like the group discussions where they could talk to each other about the projects they were working on.

But I don't have a good way to analyze those other than reading each one. I may add it someday, but it's not high on my list of things to do.