Use the attached template to calculate the discrimination of your quiz questions using Brightspace Data Sets, downloaded from Data Hub.* For your reference, we have also included a completed template using sample data.

Need to learn more about using Data Hub? Watch these video tutorials.

Have more questions? Post a comment below to leverage the knowledge of the Analytics and Data Access community.

*Available in Brightspace Core only

1. Rubric Assessment dataset, according to the rubric datasets documentation, "Returns rubric outcomes that are associated with a competency." I think I am pulling data from courses that do not have rubrics associated with competencies. Since most of our courses do not use competencies, it is a good thing I am getting data from them! But could I be losing this data because I am not supposed to get it? Please remove my anxiety, or give me the warning I need.
2. Our content-provisioning process uses master classes that are copied every semester into various "live" course offerings. I think that the RubricId is unique to a rubric in an offering. Correct?
3. What happens when there is a shared rubric is in a course, coming from another course? (I know this locks the rubric in the LE, but what happens in analytics?) And if a shared rubric is in a master course? (I don't know the answer to this one and have no real-life example test but it could come up at some point.)
4. Again with master courses and many clones in mind, I think that AssessmentId (shared between RubricAssessments and RubricAssessmentCriteria, among others) is unique to an assessment (discussion, assignment) in a course offering. Correct?
5. Does this AssessmentId appear in any other group of datasets? (I don't see it but thought I would ask for planning purposes.)
6. A course can, for example, attach a single discussion rubric to a dozen discussion topics. I see the same RubricId for each Discussion topic, and the same CriterionId for each time the criterion is assessed for various students,, so the UserID and the AssessmentId combined with the RubricID and the CriterionId provides the unique assessment morsel that differentiates each row in RubricAssessmentCriteria.
7. The Score in RubricAssessmentCriteria is the score given by the instructor vs. the default score associated with the criterion-level selected.

https://www.youtube.com/watch?v=7URqian6Zdk

For Production sites in 10.8.8, we released our first new dashboard in the Insights Portal: Adoption. This powerful new dashboard provides an org wide view of the use of Brightspace. You will be able to understand who is using Brightspace, how much and in what ways. It reports on the basics such as logins over time to tool usage by org unit for a deeper analysis.  

Learn more here:

https://community.brightspace.com/s/article/New-Insights-Adoption-Dashboard

For example, for the following outcome, the Mastery View is showing Proficient and hidden for the test student when I look at the course in D2L, yet when sorting by latest assessed date, the level is Developing and IsPublished is null.

I was going to just use the latest assessed date to pull out the level from the data sets, but obviously that wouldn't work. (I got past this by replacing nulls and ordering on both IsPublished and AssessedDate using Rank & Window partitioned by student and outcome.) I just don't understand why that row exists if that change was not actually made or how IsPublished is null again.

The D2L Product Design team wants your input on Data Hub!

What is this research about? Our goal is to understand how Data Hub is currently being used so we can build it to be as intuitive and easy as possible. To help us do that, we want to better understand your experience using Data Hub. 

How can I participate? We are aiming to hear from 100 Admins in total who are currently using Data hub. The following link will allow you to access the survey: https://survey.medallia.ca/?Data-Hub 

If you click the link and receive an error message, it is likely because we have reached the max number of responses we are aiming to collect, and the survey has closed. If you run into any other problems with accessing or completing the survey, please contact us at UX@d2l.com.  

What do I get out of the survey? As a token of our appreciation for your time and feedback there is the opportunity to receive a gift card valued at 5 USD. To receive the gift card, we will ask you to provide us your name and email, at the end of the survey. 

What happens after the survey? We can’t always follow up with you on the impact that your survey responses had on our design, but we do take the responses very seriously and ensure they have an impact on our product. 

We are just starting to use the Insight Portal. We have the following question.

We are wondering the best way to provide Program Directors with access to the Insight Portal for all of the courses in their program. The caution is that we don't want these folks to see data for courses outside of their program.  

Kind regards,

Heidi Blair

Regis University

First off, apologies if I'm posting this question to the wrong board. However, my team's working with an instructor who's dealing with the known bug involving Details Statistics and missing questions (here's the [Archived Content] article on it). Though the article's workaround was successful, the professor wants to know if the bug will persist with each new course copy. The concern feels quite valid overall, and I'm curious about it, too.

D2L reported over 1,000 attempts on a survey with one Likert question in a large class, yet the Statistics and Reports only included 300 responses. The instructor had updated the question (just fixing a typo, so no major changes) and several students' attempts warned "This question is an old version as it was modified after the quiz was taken". Their agree/disagree responses were recorded on their attempt page, yet the completion summary was blank as if they had not responded to anything and they were not counted among the 300 responses.

Even querying the datasets only resulted in 300 responses when joining Survey Attempts, Survey Question Answers, Survey User Answer Responses, and Users. I had to omit Survey Question Answers (i.e. the text of the Likert statements) in order to get all 1,000 responses. I discovered that the AnswerIDs were not included in that data set but I do not know why. I also discovered the QuestionVersionID for the old question was missing from the Question Library data set.

Is it normal that data for an older version of a question is missing from the data sets? Even though the student responses were still available? This question was created within the past 2 years. How would instructors who only have access to D2L and not the data sets go about downloading this data from D2L?

This means that if you have not updated your BDS version to 7 it will be done automatically once you received the April release.

If you are looking for details about what changed in BDS v7 check out this blog post:

https://community.brightspace.com/s/article/Motivations-behind-the-changes-in-Data-Hub-Brightspace-Data-Set-Major-Version-7-0-October-2021-Release

And if you are looking to understand how this affects your Insights Report Builder check out this blog post:

https://community.brightspace.com/s/article/Updating-Insights-Report-Builder-for-Brightspace-Data-Set-Major-Version-7

No worries - the recording is available here:

https://www.d2l.com/resources/webinars/intro-to-data-hub/

Check out the NEW 'Everything You Need to Explore Data and Analytics with Brightspace' page, a centralized resource for answers to these questions and more! 

https://community.brightspace.com/s/article/Everything-You-Need-to-Explore-Data-and-Analytics-with-Brightspace