Challenging Trends: Data, Learning and the LRS

• Home
• Blog
• Learning & Development

Chris Dunne, Digital Marketing Manager

As learning professionals and technologists, we’re no strangers to the latest industry trends, but keeping up can be something of a challenge. Adaptive Learning, Artificial Intelligence, Learning Analytics; the list goes on. But what do all of these trends actually mean for you and your learners? 

Learning Pool‘s latest webinar series, Challenging Trends was designed with the aim to:  

• Explore the origins of the most popular ideas in digital learning today  
• Compare the trends we can see to the challenges we face  
• Dissect the possibilities and potential emerging technologies  

In his role as Chief Product Officer at Learning Pool, Ben Betts joined Learning Pool’s COO, Deborah Limb to form a panel for the ‘Data, Learning and the LRS’ webinar where they discussed the growing importance of data in learning.  

The webinar included a Q&A session, with Ben and Deborah responding to many insightful questions regarding the Learning Record Store (LRS) and the xAPI. If you need more information regarding either of those things, you can download the Learning Technology Manager’s Guide to xAPI.  

If you missed the webinar, or simply want to recap, take a look at their answers below.  

How easy is it to record activity from a non xAPI source? 

“There are three basic ways of getting data into an xAPI Learning Record Store like Learning Pool LRS. The first is xAPI, that is the activity source is already outputting data in the xAPI format, this is known as ‘Plug and Play’.  

The next option is that the activity provider has an API but it doesn’t conform to the xAPI specification. This is very common. Here, we can translate between the two, dependant a bit on the sort of data being recorded. We’ve done this for many common platforms in our area already, including Skillsoft, Degreed and Yammer.  

The final method is that there is no API available at all, but we can get a CSV (spreadsheet). Here, we translate rows of data into xAPI statements using templates. This can be done via a script automatically, or the CSV can be dropped into a secure folder, or even emailed.  

The method used usually depends on the frequency of data, for example, classroom registrations might happen once a week or less, so could be emailed. Other data happens frequently and so must be in almost constant sync. Learning Pool’s LRS apps can do all this for you.” 

Can individuals self-record on an LRS? For example, “I listened to a TED talk, I want to record that as CPD.” 

“Whilst it is possible to self-record learning activity to an LRS, usually with a bookmarklet in the browser, engagement with these features is usually very low. We’ve seen examples of forms being offered on intranets or LMS’s to specifically allow folks to register external CPD to count towards annual targets. These work better, and our LXP offering also has this functionality built-in.” 

Does the unique personal record data have to match (including the hierarchy of their organization) for the data to output in the right way? 

“These is a separation in Learning Pool LRS between persona data (about people) and activity data (about learning experiences). Learning experience data is immutable, you can’t change it, it’s a log of what happened. Persona data can be edited and added to.  

What many organizations we work with find, is that they have multiple identifiers for the same person; a unique number, an email and so on. Personas in Learning Pool LRS allow you to register multiple IDs for the same person, so that when you query you can return every learning experience record for each unique identifier. This can also work with organization hierarchy data which can be recut and appended at any time.” 

How will the data take into account different nationalities/cultures? 

“Building on the concepts of ‘personas’, that is, data relating to a person that is not just a log of their learning activity, gives us an opportunity to create hypotheses based on localization data, for example, do folks in Malaysia upload more videos than those in the US? (In our experience, yes they do!) We do this by simply adding more data to an individual’s persona – most often done with a link to the HR system or similar.  

One has to be careful about biases and other issues that abound from data-led decision making, but the opportunity is there for personalization.” 

Could SCORM and the xAPI sit together in an LMS or LXP? 

“SCORM and xAPI can absolutely co-exist, and often they do. I suggest they are different branches of the same evolutionary tree; the xAPI does not mean the end of SCORM. xAPI won’t do a host of things (natively) that SCORM will do, typically around the sequencing and packaging of content. It’s still hard to get an LMS that can securely launch and track xAPI content, although Learning Pool Platform does it.  

But equally, pain exists the other way around. SCORM lacks granularity in the data, it doesn’t suit many more social or informal means of learning and it’s not practical for more micro-content; authoring or consumption.  

Ultimately, I believe you should be looking to create a single source of data for your analysis needs. And this might mean initially storing SCORM data but transforming it to xAPI. As the tech evolves you may find SCORM becoming more redundant but I wouldn’t fancy that particular branch is going to die off any time soon!” 

If it’s Open Source, could internal web developers use that to pull information from Sharepoint without the plugins? 

“Learning Pool LRS is available Open Source, so a competent developer can install, modify and use it without change. That same developer could readily create a plugin to transform Sharepoint data to xAPI and store it in their Learning Locker.  

However, this will cost them both time and money. So it might be more cost-effective to just buy the plugin from us. Generally speaking, the plugins we’ve made cost something and you need to need to be on the Enterprise version of Learning Pool LRS to use them.” 

The dashboards look great, but can other BI tools connect in Tableau, for example? 

“Yes, we have a BI connector tool, available as an app for Learning Pool LRS, that can connect data in real-time to various BI tools, Tableau and PowerBI included. It’s common for organizations to connect their LRS to a BI tool for deeper analysis.” 

Can you easily export data from the LRS to plug into Azure or AWL ML services, for example? 

“We offer a service to connect an AWS Kinesis stream to the Learning Record Store, allowing you to effectively firehose any other AWS service. Azure support is a little more limited right now but we are actively working in this area.” 

Are there any restrictions on the type of data that can be aggregated? 

“The LRS can literally take anything as long as it conforms to the xAPI specification. We once had it hooked up to our Principal Consultant’s Fitbit and had to aggregate up from every step as it was getting a bit annoying!” 

Can the LRS data be accessed by other applications, e.g. custom dashboards, via a REST API or some other mechanism? 

“Everything stored in Learning Pool LRS can be accessed via our APIs with the correct authentication. You can check out the docs for the REST models and others here.” 

Are there new privacy implications about an organization collecting and analyzing far more data about users than it did previously? 

“There are significant implications for storing more personal data. I believe one of the first things an L&D/HR function should do is come up with a code of conduct for how it intends to use learning experience data. This can be seen as your ‘north star’ for everything that goes on before, and should be different to T&Cs, which are a legal document.  

A code of conduct should be no more than a page, be written in plain English and should sketch the boundaries of what an organization will/won’t do with learning analytics data.  

See this primer from education as a good starting point.” 

What role do you see the LRS/data serving in recommendations to individuals based on previous actions/interests? 

“Anytime I hear an organization wanting to deploy smart recommendations for learning resources, I immediately think ‘where’s the data’? To do recommendations effectively takes a pretty big amount of data to overcome the ‘cold start’ problem, so getting going here without some legacy in data is a challenge.  

This is why I think it’s important to start collecting data today in a standard format, even if you think recommendations won’t be a thing for a few years to come. When the time is right, you’ll want the data.” 

If you’d like to learn more about Learning Pool LRS, the award-winning Learning Record Store, arrange a demo.