Data is important, but it is the stories it helps to tell that make a difference.

Take a look at some of the stories that can be told with the Tableau tool: Examples of Business Intelligence dashboards
Tableau Viz Gallery | Tableau Software

The plan is to export a csv file from Snowflake and import it into a Tableau account. Please remember that the free accounts are public, even if reports are unlisted, they can still be downloaded by anyone. So, no FERPA data should be included. If you need to share student names, make sure to license a Tableau server.

Step 1. Find the SQL and run it.

You can explore SQL scenarios and code contributed by various Blackboard community members. Let’s play with Carlos Guevara SQL about Collaborate. The snippet he posted is nice, but perhaps you’d like to see how your usage grew over time. Let’s extract the session date, and then group by year. This SQL should run for most Blackboard Data users:

 

select
  count(distinct a.person_id) as student_count,
  count(distinct col.room_id) as room_count,
  count(col.id) as session_count,
count(distinct c.id) as course_count,
        month(col.start_time) || '/' ||  year(col.start_time) as session_date
      from cdm_map.course_room map, cdm_lms.course c,
      cdm_clb.session col
      left join cdm_clb.attendance a on a.session_id = col.id
where
      c.id = map.lms_course_id and map.clb_room_id = col.room_id
      and col.start_time > '2017-01-01'
      group by session_date
      order by session_date

 

Sample data output for this SQL:

Screen Shot 2020-07-07 at 6.14.22 AM

Step 2. Download the CSV file.

Notice the download icon next to the Copy button. This will allow you to download the data as CSV.

Screen Shot 2020-07-07 at 6.15.23 AM

I’m calling my data export collab_data (.csv will be added automatically). This file name is important, because once you cannot it to Tableau dashboard you will set the name as a data source, so you should keep the data files unique.

Step 3. Upload the data as a data source to Tableau.

Earlier I setup a free account on https://public.tableau.com and downloaded the desktop app. I’m running it on a Mac. Each time you save a dashboard it will build a web report starting at the exact view you are looking at. Mastering Tableau will take a little bit, but it’s worth it.

A. Start the desktop Tableau app.

B. Select Text file as a data source.

 

Screen Shot 2020-07-07 at 6.16.16 AM

C. Make sure to change data type of the session_date from STRING to DATE. It will allow for the right kind of reports (line graph, filtering by year etc).

Screen Shot 2020-07-07 at 6.16.24 AM

D. Create your sheet, dashboard, and story. When you save it will publish. You can hide your report after publishing it so it is unlisted.

Once the data source is created use the “Sheet 1” button at the bottom left of the screen. You can download my demo report to get you started. YouTube is full of tutorials. The reports will help you visualize your data and tell your story.

This is a sample live report:

https://public.tableau.com/profile/szymon.machajewski#!/vizhome/DemoBlackboardCollaborateReport/Story1

 

Screen Shot 2020-07-07 at 6.17.54 AM

This sample report is very simple. However, imagine seeing a department list, then drill down to courses of that department, then view student activity, and drill down right to individual student sessions. That’s powerful.

PS.
If your course_id is structured as: 2019.spring.ma.101.12345 you can use this SQL to run term/year groups:

select
  count(distinct a.person_id) as student_count,
  count(distinct col.room_id) as room_count,
  count(col.id) as session_count,
 
count(distinct c.id) as course_count,
  split_part(c.course_number,'.',1) as Year,
  upper(split_part(c.course_number,'.',2)) as Term
from cdm_map.course_room map, cdm_lms.course c,
cdm_clb.session col
left join cdm_clb.attendance a on a.session_id = col.id
where
c.id = map.lms_course_id and map.clb_room_id = col.room_id
group by Year, Term
having try_to_number(Year) > 2015 and 
Term in ('FALL', 'SPRING', 'SUMMER')
order by year, term




Screen Shot 2020-07-07 at 7.46.52 AM