Activity 3 – Week 12 (Analysing Learning Analytics – Educators/Instructional Designers)
This experience is designed to create an analysis of the current market for Learning Analytics Ventures. Its purpose is to help you understand critical information in the analysis, design and evaluation of a social/learning analytics venture. Multiple roles are presented: students, educators, instructional designers, venture analysts, entrepreneurs, and investors.
You may respond to one or more of the categories as desired.
This Blog is dedicated for your response as Educators/Instructional Designers.
How would learning analytics/social analytics be helpful to you?
How would you use learning analytics within your instructional design or to support your classroom/teaching practice?
These questions are prompts. You may use them or respond in any other way.
To respond, simply reply in the comments section below this post.
To respond to other roles go to:
To review or for further reading visit our Presentation.
Posted in: Week 12: Social Analytics
Kristopher 9:43 am on November 23, 2011 Permalink | Log in to Reply
Learning analytics would be helpful as they provide a huge amount of information in regards to how the learner is interacting with the internet. There are of course limitations, the biggest of which is falling into the trap that that type of data collection can be used exclusively instead of combining that type of information gathering with other assessments.
As an instructional designer, I would use this type of information to note what learners are doing, so that I could cross reference with what learners should be doing. This would give me a sense of where the content/activities are lacking relevance or excitement.
In addition, analytics can help us to understand the context in which the learner is operating. By understanding the context, we can improve the relevance and engagement of the learner with the content. Again, the same pitfall (that we may be relying to heavily on one set of data) exists.
Kristopher
Julie S 11:19 am on November 23, 2011 Permalink | Log in to Reply
I did an experiment with learning analytics in my ETEC 511 final report. I used the Social Network Analysis (SNA) software to analyze the course discussion groups and found the results very interesting and useful from the perspective of future instructional designs for online training for the workplace.
During my investigation of this technology I found research by Shen et al (2008) that directly related a student’s perception of community on an online environment with their interactions in the discussion forum. The social network analysis (SNA) was seen to be a valuable tool for exposing the interactions. Shen et al’s key argument is that of the ability to determine optimal learning environments, density of network, centrality (# of ties, how inter-related the nodes are, how close to the centre). I think that the effective use of the technology requires the active involvement of the instructor to see if learners are either lurking or dominating in discussions and to take proactive action to encourage discussions where everyone participates.
The challenge with the technology was that it could only be used within controlled discussion forums and didn’t take into account alternative communication channels or social networking applications. So, I agree with Kristopher’s perspective that it shouldn’t be used exclusively as it can only provide part of the picture. However, it still could help the instructor from a monitoring perspective.
Reference:
Shen, Demei, Nuankhieo, Piyanan, Huang, Xinxin, Amelung, Christopher, Laffey, James. (2008). Using Social Network Analysis to Understand Sense of Community in an Online Learning Environment. Educational Computing research. Vol. 39(1). 17-36, 2008.
mcquaid 5:31 pm on November 23, 2011 Permalink | Log in to Reply
It would be neat / useful if, in a course that was being monitored this way, a student could input the sites / channels / avenues they would be communicating in so that everything that “counted” was recognized.
jarvise 6:46 am on November 24, 2011 Permalink | Log in to Reply
As an educator, I have been reflecting on the use of social analytics to personalize and target instruction. I’ve been thinking a lot about the TED talk presented, and struggling with the benefits of personalization on one hand, and the drawbacks of restricting query results to those areas that have been identified as important. There is an ethical issue to targeting a person’s learning preferences based on what they have done in the past. If a tool is targeting information and activities to our past behaviour (I keep hearing Dr.Phil’s voice: “the best predictor of future behavior is past behavior…”) then how do we have the opportunity to change and adjust our habits (in other words, to learn new behaviors)? When working within the realm of education, we often focus on learning about our students preferred learning styles, and making sure that our teaching incorporates activities that target them. However, within a classroom, there are often so many learning styles that all students would be getting exposed to different avenues for learning at some point – not just the ones they like the best at the time. If hidden analytics are individualizing an activity, then how do we have the opportunity to consider other ways of doing things; to gain exposure to how someone else may approach an exercise? When it comes to tracking a whole site (or a whole course), and using the information to improve the usefulness of it, I don’t see the dilemma. When it comes to targeting people, though, it seems restrictive. Sometimes I love that Amazon can predict what I want, but other times, I’d just like to see a bird’s eye perspective on what is out there and what is popular (and not just popular with people like me…).
Emily
Allie 3:03 pm on November 24, 2011 Permalink | Log in to Reply
Thanks for your thoughtful response, Emily! It seems to me, and please correct me if I’m paraphrasing you incorrectly, that your primary concern with social/learning analytics as used to guide personalized learning plans (as with School of One) is that it risks being too individualizing? Like, a classroom would be full of, say, 30 little learning silos but no social learning happening between them?
I think that’s a really interesting point and critique of personalized or 1:1 learning – especially since much online learning environments seem to depend on the sharing and social building of knowledge.
David William Price 7:42 am on November 25, 2011 Permalink | Log in to Reply
The TED talk made a great point that may be getting lost. It’s not that Google is doing anything differently than we already have. We’ve always had curation. There’s too much information to understand and we need trusted curators. Even within a classroom, we tend to rely on the professor as a curator. We are willing to pay vast amounts of money to have people curate knowledge for us… essentially that’s what role specialization is all about. We take recommendations from friends and family instead of doing research ourselves. The amount of information available is staggering.
Re “learning styles”, try Googling “neuromyths learning styles”. It’s an example of an educational “trend” with blinders on and the trend seems perfectly related to the concept of getting lost in a filter bubble. People are still talking about left-brain and right-brain activities long after the concept has been thrown out the window in neuroscience.
Allie 12:39 pm on November 25, 2011 Permalink | Log in to Reply
indeed, human perception itself is a curatorial act of sorts as what we actually see when we’re looking at something is filtered in culturally specific ways.
Deb Kim 10:18 pm on November 24, 2011 Permalink | Log in to Reply
Social analytics could help me identify students who are disconnected from social network. I can also identify high and low performing students. I can see who has been interacting with whom. By identifying them before marking their work, I can plan interventions so that I can help them increase their participation.
Moreover, I can monitor students’ posts (e.g. the total number of posts per student). This can help me monitor their participation level.
Deb
David William Price 7:57 am on November 25, 2011 Permalink | Log in to Reply
If you do Google searches for ideas, you don’t get balanced results. Say you want a review on a product you’re considering. If you Google for “product review” you often get a lot of highly positive results. When I was shopping for a car many years ago, I found you could get endless glowing reviews of new cars and their features and capabilities, even from reputable car magazines.
So what’s missing? A simple word: “problems”. Try doing the same searches for reviews but instead of “product review” search for “product problems”. Suddenly that fabulous car you’ve been reading great things about doesn’t look so good at all. Suddenly that amazing computer monitor seems like an overpriced, flaky chunk of frustration.
We get absorbed in hype and trends. People honestly chase optimism. It doesn’t mean they are necessarily malicious, calculating or controlling.
How could analytics help us? Analytics could be a performance support to ensure better critical thinking and more balanced design and teaching by tracking “problems” and “issues” and “controversy” even while they are not trending large in pure numbers. What if analytics were a sidebar on our search engines, tracking the pitfalls and challenges and concerns behind our searches? It’s like a friend who taps you on the shoulder and says, “I overheard you talking about X. X has a lot of good features, but did you know it has pitfalls A and B?”
How do we apply this to learning? I think the whole left brain, right brain, visual-auditory-kinesthetic learning styles trends are a good example. A number of articles came out a few years ago that basically said science does not support these concepts. Many teachers and instructional designers use techniques that have no scientific basis or worse have been proved ineffective. Analytics could provide an ongoing balanced view showing trends in scientific research vs. popular trends. Popular trends tend to trail science by years if not decades. Perhaps analytics could provide corrective information by plotting the scientific directions vs. the popular direction and you could see divergence much faster and follow up on it.
jarvise 6:38 pm on November 25, 2011 Permalink | Log in to Reply
Hi David,
Great points. I love your tracking ‘issues’ idea. This would be very valuable. The only issue I see arising from it is how strongly we value a negative review over a positive one. When I’m reading reviews from 10 people about a pair of snowshoes, even if it has been rated 4 and a half stars overall, I’m still hesitant to buy them if there is one bad review. I can’t let it go. You’re right that we chase optimism (like a herd, really), but we get swayed by pessimism very easily as well. We are so swayed by anecdotal evidence as it is – it seems like we may be stoking the fire by providing such ready access.
How many of us got flu shots this year? Consider the reasons.
Emily