CU Denver Shares Canvas Data/Tableau

[kenneth_rogers]

My first introduction to canvas data‌ and tableau‌ came at the hands of both  @brenden_goetz ‌ and  @alexander_kark1 ‌. These topics have been mentioned around my institution, and they have picked up steam as of late, so I figured what better way to get some additional information than in this session.

The presentation focused on their mobile learning report. As a member of Canvas Mobile Users‌, this is near and dear to my heart. I loved how they focused on how a student was producing data vs. consuming data. With this information they will now have actionable data to inform course and mobile design. But at a deeper level, what else can you do with the data?

I had a few key takeaways...

• Because CU Denver has multiple seats of Tableau, they have reports for more than just Instructional Design purposes. IT can have reports to see when students are accessing Canvas (to know when to do other maintenance). Or you can begin to look at when students are taking exams most frequently so you can work with instructors and say "Maybe Sunday at 7:00 isn't a good deadline."
• Actually look at the data and ensure it makes sense. CU Denver saw an exceptionally high activity in "Files" in a course that had no course files for students to access. However, when they looked at the course, they noticed the course had design elements from the files area, which was causing the activity. Make sure you double check everything!
• The goal is really for a collaborative Canvas Data Community. They have Instructure host their Canvas Data with Amazon Redshift, and Brenden has already provided a lengthy document here in the Community for reference. Check out https://community.canvaslms.com/thread/13718-device-usage-in-canvas-data-with-tableau for reference.

[joseph_allen]

Hi Kenneth,

I think one of my big focus areas at the recent #instcon 2017‌ was to better understand Canvas Data and Tableau.  As you know there were many sessions dedicated to various aspects of the topic.  I think these tools are very powerful although I'm not sure there is broad understanding on exactly how to use them.  Our school district plans to dedicate significant resources to this topic during this school year to really mine data, understand user behavior, establish some accountability, and evaluate the return on investment in various areas of Canvas.  I look forward to comparing notes with you and others on how things are going with Canvas Data.  

[rseilham]

 @brenden_goetz ‌ &  @alexander_kark1 ‌

I'm really sorry I missed this session. I can't wait to watch the recording later on.  I've worked with our database engineer to pull data for mobile in Canvas Data. A lot of it was finding the proper user agents, which took a lot of research because it's changed over time. There is also a lot of noise, particularly with the Android app. I'm not sure if this is helpful, but if you haven't, check out this CanvasLIVE presentation I did last year:

This video is currently being processed. Please try again in a few minutes.

‌. 

I'll also leave the user agents below: 

iOS (Red App)

• Canvas - ex: Canvas/3.14.0 (iPhone; iOS 9.3.2; Scale/2.00))
• iCanvas - ex: iCanvas/3.13.2 (iPhone; iOS 9.3.2; Scale/2.00))
• Canvas/com.instructure.icanvas - ex: Canvas/com.instructure.icanvas (1035; OS Version 9.3.2 (Build 13F69))
• CanvasKit - ex: CanvasKit/1.0

Android (Red App)

• candroid - ex: candroid/4.5.0 (122)

SpeedGrader

• SpeedGrader - ex: SpeedGrader/226 (iPad; iOS 8.1.2; Scale/2.00)

Canvas Polls

• polling - ex: polling/1.0 (iPhone; iOS 8.3; Scale/2.00)

MagicMarker

• MagicMarker - ex: MagicMarker/1.1 (iPhone; iOS 8.1.2; Scale/2.00)

Canvas Teacher iOS

• Teacher - ex: Teacher/0.3 (iPhone; iOS 11.0; Scale/3.00)

Canvas Teacher Android

• candroid - This will be changing in the next version to something else.

[brenden_goetz]

Thanks for sharing those user agents  @rseilham ‌! I did watch your Canvas LIVE! webinar last fall, but need to circle back and re-watch it as we keep digging into the device usage. 

What was your research process like for tying user agents to specific devices or platforms? We have been doing a combo of "user simulation" (shout out to  @abiller ‌ to coming up with a good term for that) to see how records show up in the requests table when we know what device and platform were being used, and general web research to use existing knowledge about user agents (e.g. Mozilla Developer Network). Just curious if you're sniffing around in the same way. 

[rseilham]

Hi  @brenden_goetz ‌, 

I love the term "user simulation" - that's awesome! Honestly, you are probably doing much more than us. I am lucky to have a database analyst that knows so much more than I do. I asked him his take on your question and he said: 

One of the standard ways to get device and platform information from the user agent string of both web and mobile/tablet is by using the npm package platform

An example of it would be like
var info = platform.parse('Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7.2; en; rv:2.0) Gecko/20100101 Firefox/4.0 Opera 11.52');
info.name; // 'Opera'
info.version; // '11.52'
etc
User agent strings from mobile/tablet devices give out similar information
platform.name; // 'Safari'
platform.version; // '5.1'
platform.product; // 'iPad'
platform.manufacturer; // 'Apple'
platform.layout; // 'WebKit'
platform.os; // 'iOS 5.0'
platform.description; // 'Safari 5.1 on Apple iPad (iOS 5.0)'

Link: https://www.npmjs.com/package/platform .

Or you can go in a creative path of building regular expressions to parse the information.

Let me know if this helps!