Dr. George Siemens is professor at the University of Texas at Arlington and director of the Centre for Change and Complexity in Learning (C3L) at the University of South Australia. He’s an internationally known author and speaker who has delivered keynote addresses in more than 35 countries. He’s also a researcher and theorist in the field of learning, knowledge management, and technology.
George is recognized for developing the learning theory of connectivism as well as for his pioneering work in learning analytics and the development of massive open online courses (MOOCs). And he’s continued to develop and deliver MOOCs, including one now on helping instructors new to online make the pivot necessitated by COVID-19.
In this episode of the Leading Learning Podcast, Celisa talks with George about connectivism—the theoretical framework he developed for understanding learning in a digital age. They also discuss the strengths and limitations of MOOCs as well as the potential positive and negative impacts of learning analytics with practical insight on how to begin using them in your organization.
To tune in, just click below. To make sure you catch all of the future episodes, be sure to subscribe by RSS, Apple Podcasts, Spotify, Stitcher Radio, iHeartRadio, PodBean, or any podcatcher service you may use (e.g., Overcast). And, if you like the podcast, be sure to give it a tweet!
Listen to the Show
Read the Show Notes
[00:18] – A preview of what will be covered in this episode where Celisa interviews Dr. George Siemens.
[02:06] – We aren’t going to offer reflection questions for this episode as a nod to George’s emphasis on self-regulation. He argues that one of the most urgent skill sets that we need in this current open learning landscape is how to set goals, how to develop learning strategies, and how to monitor progress towards those goals and the effectiveness of those strategies. Those goals and strategies and measurement can’t—or at least shouldn’t—be handed to learners if the learning is to be effective and deep.
That, of course, dovetails with what we’ve talked and written a lot—namely that learning is effortful. While we have a growing number of tools and technologies—and even more effective tools and technologies (thanks to things like AI and learning analytics)—ease can be a bad thing in learning, where experimentation and effort are key.
So this time around the reflection question is a DIY project—craft your own question or questions after you listen to what George has to say. And feel free to share those in the comments section at the bottom of this post.
[03:46] – Introduction to George.
The Idea Behind Connectivism
[04:33] – Let’s start with connectivism, which dates back to the early 2000. Briefly what is connectivism, and why did you see the need to posit it?
George shares how the current environment is a bit different from when the idea of connectivism first formed over 15 years ago. At that time, he was part of a group that was actively involved in using technology to connect, share, and collaborate globally. And he this felt like it was a structurally different experience than what he had been informed learning was.
George talks about how there was a sense that the shape of the network and people he was interacting with was distinct and unique and provided a very different set of affordances and opportunities. He found that his engagement with that network in many ways is what made him intelligent—or capable of being intelligent.
So the core idea that he posited was that when we learn, we’re essentially involved in a process of pruning, forming, and developing our networks. The argument at that stage was that these networks occur at three distinct levels:
- There’s a biological underpinning of learning. We have neurons connecting/forming when we learn a new concept, recognize a person, try to recall something, etc.—we’re activating existing networks. So the network learning lens in connectivism addresses that at an absolute biological level.
- This is much of what we do in our school systems/formal learning today. You learn a new idea, relate it to other ideas, you engage with concepts and see if they fit together, and you eventually develop an understanding of a domain that allows you to get a degree or formal recognition. How we relate ideas to other ideas is essentially what is meant by conceptual networks of learning.
- External (Social networks/systems). This is probably most relevant to people in the professional learning space. The social networks and systems are part of our capability to be intelligent. And increasingly, that’s starting to include cognitive agents such as apps on our phone, and increasingly look like it might be artificial intelligent agents that help us work through navigating information.
How Connectivism Has Changed
[08:26] – So connectivism is at least 15 years old, at this point. What’s changed in that decade and a half since you originated the theory? Does connectivism still apply to how learning happens? Has there been the need for tweaks or major changes?
George thinks it absolutely applies and, if anything, it applies perhaps more than it did at the time that he first put it forward. In fact, he recently put in a book submission on connectivism. The interest and recognition that this is a way of engaging with others in a way that can impact us across the duration of our lives seems to still be quite an essential area of focus.
So the core ideas are still there that learning is a process of being network and connected (and that occurs at the three levels outlined above).
George says there’s a growing importance—especially when we’re dealing with complex problems or phenomena—that the learning or knowledge acquisition is not being done by an individual.
More and more everything that we do is a function of teams and networks. And we hold pieces of information within that network.
And if we’re trying to solve a complex challenge or confront a complex opportunity, we rely on activating those networks and systems. If anything, the appeal of the network in complex system becomes more significant, the need for us to devote attention and effort to how we connect to others and how we connect concepts becomes increasingly more significant.
George notes the primary tweaks have just been the growing recognition around AI in these external networks. And also the greater need for coordination mechanisms as systems become more and more complex.
Strengths and Limitations of the MOOC Format
[11:06] – You pioneered massive open online courses, and I know you continue to be involved in MOOCs—you’re teaching a new one now, as we’re talking, focused on helping instructors new to online make the pivot necessitated by COVID-19. What do you see as the strong points of the MOOC as a format, and what do you see as its limitations?
As a format, George says he thinks MOOCs represent a cycle of change that has been going on for decades (probably almost a century). And that change is that we are more and more required to activate knowledge and learning in our daily lives beyond formal schooling.
He discusses how 30-40 years ago, the view was that you do K-12 schooling, maybe a four-year degree, and then you get a job and work. But now we’re spending more of our career learning informally than we are formally.
George talks about how if you graduated at 25 and you’re now 50, you’ve had to rely on perpetual, ongoing learning needs—almost the idea of learning as a way of being. So what MOOCs represent is a recognition that we’ve had a demand side increase in learning needs that runs 50-70 years. But we really haven’t had a structural supply side increase in providing the needs that those learners have.
As opposed to the university model that still assumes the centrality of the traditional university sector where the learner adjusts her life toward that model, George says MOOCs represent a recognition that that’s not how learning works anymore.
Learning is perpetual and ongoing. And the learners themselves need the ability to have their needs met while they’re having a family or engaged in other various opportunities.
It’s primarily the output of that experience and that’s what MOOCs do very well—they provide flexible, accessible learning opportunities to individuals who are living complex and complicated lives.
What MOOCs don’t do well is provide the traditional support structures that help learners to be successful. George shares how they’ve recognized the value of a schedule to help students stay on task.
There’s a far greater need for self-regulation on the part of individual students.
One of the reasons he says MOOCs have the high failure/dropout rates is based (at least partially) on the fact that those support infrastructures aren’t in place.
Another aspect may just be that people don’t necessarily want to complete the MOOC but want discreet knowledge to solve a particular task.
But more of the challenge with MOOCs is that we don’t know how to support and promote self-regulated learning in these online/distance environments at the same level that we’ve built within our university campuses.
[15:53] – You’ve been involved in MOOCs for a long time so in terms of their relevance, I think I’m hearing from what you’ve already said that you imagine them continuing to be very relevant going forward. Is that correct?
George says they are absolutely still relevant and notes some statistics that show there are roughly 200 million people who’ve signed up and been involved in a MOOC.
He points out that some of the most popular MOOCs on platforms like edX and Coursera have an interesting dichotomy. On the one hand, it’s heavily based towards computer science and programming/data science skills.
But on the flipside, some of the most popular courses are on happiness, positive psychology, and well-being. So you have that interesting mix where you have both learning for employment and related opportunities, but also a deep interest on the part of many students for just a better quality of life.
Going forward, George says universities haven’t built that pipeline to meet the needs of learners across the duration of their lives. So concepts like MOOCs or related platforms such as LinkedIn Learning or Kahn Academy will only continue to grow in attention and relevance.
And even though MOOCs have been around for about 8 years now, they’re still seeing rapid growth. And if anything, the current crisis with COVID is driving a lot more attention to recreational or personal interest learning in the form of MOOCs and MOOC platforms.
Learning Analytics – The Good
[18:25] – Learning analytics is an area of focus for you. What do you see as the most exciting potential for how learning analytics are being used—or might be used?
George describes how learning analytics provide two significant points of value to learning professionals, corporate learning, universities, and any other kind of related learning setting where learning is structurally planned formally taught, and then assessed and evaluated.
On the one hand, he says it will and is providing advances to how we understand learning. At its core learning analytics is a learning sciences psychology of learning contributing player.
And what that means is we understand—through the actions that individuals engage in as they’re learning an online setting or learning with digital technologies—is that we gain understanding into self-regulatory habits. We gain understanding in terms of the behavioral clusters that students fit into when they’re involved with learning a new idea or when they get frustrated or overwhelmed.
In that regard, George thinks learning analytics has significant potential to contribute to our understanding of how learning occurs, especially in digital settings.
A second aspect that he discusses (that may be more relevant to learning businesses and universities) is that as we understand learning and as we understand the points of drop off for students (such as they lose interest, get overwhelmed or confused) we have a mechanism for near real-time feedback to everyone involved in the system.
So you have indications that students drop off at a certain point in the course or professional learning program and George talks about how you can pinpoint what the issue is.
And these two trajectories—the contribution to basic science in our understanding of learning, as well as very practical intervention activities for scaffolding support, guiding learners to better self-regulatory capabilities influencing the learning design process, influencing how teach, etc.—these are the two strands that learning analytics contributes to most heavily in both formal education, but also in corporate learning or informal learning areas.
Learning Analytics – The (Potential) Bad
[21:28] – Do you have any reservations or concerns around how learning analytics might be used? Any potential for the flaws in the systems or those two streams you were talking about and how that data could be misused potentially?
George acknowledges there absolutely is that possibility and we have that anytime there’s data and data that’s often fragmented across different platforms. There are enough illustrations from social media where it’s been misused, intentionally abused, used inappropriately, etc. And that’s something in the learning analytics community they’ve had as an ongoing conversation for almost a decade now.
The questions they are asking specifically are, what’s the value tradeoff when you provide your data to an institution? What are you expecting them to do with the data reasonably and what is it that they can do to help you become a better learner by virtue of having data about you available? He then shares some examples of how we all benefit from it.
But where things change is when that starts to get at some of our core learning needs and learning activities. To illustrate this, George stresses one of the most urgent skillsets that we need in this kind of open learning setting is how to set goals, develop learning strategies, and how to monitor our success and progress towards those goals and the effectiveness of our strategies. And that fits into this umbrella of self-regulated learning.
My concern is that when we have systems that nudge us, prod us, and move us forward, that we fail to develop those self-regulatory capabilities. Quite often the only way through to effective learning is through internal struggle, confusion, extra time and effort, and finding strategies that don’t work.
So his biggest concern about data usage is that it will help take away and under develop core needs to be in charge of our own goal-setting and planning, namely self-regulation.
There’s another area he adds that might be more of a concern for others, which is just the ethical use of that data and the privacy around it. He says it’s important to recognize that in many cases data helps make existing biases and inequalities in the system more explicit. As a result of that it can perpetuate those biases in a way that is harmful and continues to be harmful for everybody involved.
So that’s an area of focus that’s more systemic in how we are developing our models and systems that are biased against populations that may already have been underserved by our existing learning systems.
A Practical Approach to Learning Analytics
[26:39] – For our listeners, leaders and aspiring leaders working in lifelong learning, and folks who may not have access to tons of data and resources, what do you see as a practical approach to learning analytics or first steps to take? What should they be considering or doing?
George discusses how on the one hand, there is a need for people with data expertise, the necessary software or analytics skills, and the right mindsets to be able to ask provocative, grounded questions. They should also be able to verify that the results you come up with reflect proper design or data science methods that don’t misdirect the truth or the insight that’s in the data. He says the institutional skillsets can’t be overstated in terms of how urgent they are.
The second set of skills that George thinks are important are around the topic of how to understand what matters in learning—meaning you need to have a team or somebody with awareness of learning processes and what the data you have access to actually tells you. So you need the technical skills and you also need the learning-based conceptual skills.
In terms of how you get started, George references The Handbook of Learning Analytics (free PDF download) from the Society for Learning Analytics Research (SoLAR). He says this is a great place to get started to look at what is possible with analytics in learning processes. And that may spur some interest or insight into what an organization might be able to do.
George adds that they’ve also recently received approval through University of Texas at Arlington for a Master of Science in learning analytics. That’s part of a growing number of master’s programs that have either a stream in learning analytics or a focus in data and educational processes.
Aside from this, George recommends that the best way to get started is literally just to start. There are a lot of intro courses on edX and Coursera, professional online communities, the International Conference on Learning Analytics & Knowledge, and he notes that there are a group of related tools that have become much easier to use.
George says to look for the capability question—do you have the capacity internally to be able to begin working with learning data to solve learning related challenges? If not, you can utilize all of the above-mentioned resources. But he encourages people just to get started in some form.
On the Horizon
[32:00] – What do you think are the biggest challenges and opportunities facing those providing lifelong learning in the next five years or so?
George discusses how he looks at the five-year cycle and how his current interest is on human and artificial cognition—how we work together with our agents that are increasingly exhibiting intelligence type of behavior.
I think our future is one of human and artificial cognition—one where we’re going to be doing knowledge work together with a technology and that technology is going to have a certain type of agency. And many organizations aren’t quite aware of how quickly that’s coming and how quickly that’s going to make an impact on our educational processes.
He points out that we already experience it because we use it in our daily lives. And those systems that have largely been supporting human cognition are starting to flip—meaning they are starting to become a part of our cognition, to become a thinker with us.
That’s a structural change because we’re starting to offload certain cognitive capabilities to this system. We’re starting to offload some of the organizational aspects, the recommendation aspect, and the low-level cognitive tasks are being pushed onto these growing range of automated systems (something George says will only continue to increase).
If more and more of the work is being done in an automated way or more and more of the work involves an agent supplementing human cognition, then what becomes of human cognition? What do we start to do that perhaps we haven’t been doing as well in the past?
George thinks one area that’s of critical importance (for learning personnel in both universities and corporate sectors) is to figure out what it looks like when we do knowledge work with agents that aren’t human. And what is that longer-term implication, what are the skills that we need, and what are the things that we’re happy to pass off to other systems?
[36:10] – What is one of the most powerful learning experiences you’ve been involved in, as an adult, since finishing your formal education?
George talks about how we always think of learning as being additive in a way. But sometimes he thinks pruning is a really good form of learning as well. And a lot of his interest has recently been around how sense making relates to learning.
But the biggest learning experience he’s had over the last 10-15 years is actually around error correction, or a negative type of learning outcome. For example, George discusses how he used to be an adamant proponent of the incredible value of openness and transparency.
However, he’s discovered that constraints are a type of freedom—social media and platforms like Twitter have become toxic in certain communities and they are actually detrimental to learning in some cases.
So he’s finding there is value to high coherence and high trust closed spaces for learning. Not everything needs to be open and have an opinion (the way he used to think).
[39:35] – How to connect with George and/or learn more:
[40:32] – Wrap-Up
- We are not including reflection questions this go-round but we encourage you to reflect on the conversation and see if there are any questions that seems worth you asking of your learning business based on what George shared (and feel free to share in the comments below).
If you are getting value from the Leading Learning podcast, be sure to subscribe by RSS, Apple Podcasts, Spotify, Stitcher Radio, iHeartRadio, PodBean, or any podcatcher service you may use (e.g., Overcast) as it helps us get some data on the impact of what we’re doing.
We’d also appreciate if you give us a rating on Apple Podcasts by going to https://www.leadinglearning.com/apple. We personally appreciate your rating and review, but more importantly reviews and ratings play a big role in helping the podcast show up when people search for content on leading a learning business.
Finally, consider following us and sharing the good word about Leading Learning. You can find us on Twitter, Facebook, and LinkedIn. We also encourage you to use the hashtag #leadinglearning on each of those channels. However you do it, please do follow us and help spread the word about Leading Learning.
[42:27] – Sign off
- Tapping Into Educational Data Mining and Learning Analytics with Ryan Baker
- Learning Engineering and Data Analytics with Dr. Ellen Wagner
- Leveraging MOOCs for Professional Development and Continuing Education (article with resources for anyone interested in the potential of MOOCs for professional development)