OLDaily

Rewilding Your Attention
Aaron Davis, 2021/09/09

It's ironic that the source of this discussion on "going beyond the ‘inner ring of the internet‘" is "a contributing writer for the New York Times Magazine, a columnist for Wired and Smithsonian magazines, and a regular contributor to Mother Jones." This - and the Medium blog it comes from - is exactly the opposite of "stepping away from the algorithmic feeds of Big Tech." Here's the thing: most people think their internet community of such-and-such is the internet community of such-and-such, especially when it contains well-read journalists, MIT fellows, and other Persons of Interest. They rarely look beyond their immediate circles of Twitter friends and news sources. Actually rewilding your attention means eschewing the popular, getting your hands dirty with real work, and reading the people nobody else reads. Image: photo by me. 152 views.

Web: [Direct Link] [This Post]

Decision-making in uncertain times: what recent research into online learning indicates
Tony Bates, Online learning and distance education resources, 2021/09/09

To be clear, the 'recent research' referenced here is a survey of higher education institution administrations, and not faculty or staff, and certainly not anyone outside the institutions. So there's some interesting discussion here, but it's one sided. This is especially evident by the small (tiny, even) minority of respondents who listed "addressing inequalities" and "student learning outcomes" as significant challenges. Sure, it's important to have data - but having the right data is even more important. And I'm not sure the leaders in the board rooms are listening to - or providing - the right data. This will become very evident when institutions nationwide begin to experience financial stress.

Web: [Direct Link] [This Post]

Personalized Online Learning: Context Driven Massive Open Online Courses
Benmedakhene Nadira, Derdour Makhlouf, Mohamed Amroune, International Journal of Web-Based Learning and Teaching Technologies, 2021/09/09

This paper draws on efforts to combine the cMOOC and the xMOOC (in, for example, the ahMOOC and persua2mooc) and presents "a personalized model (CD-MOOC) that adapts to a learner-based approach focusing on containing learners’ dropout phenomenon." Like previous efforts, I think that this model attempts to achieve the effect of the cMOOC through personalization, whereby a centralized database of individual properties, activities and preferences is created, which is then used by an adaptive learning engine (in the current case, "three criteria were used; recommend different resources with the same efficiency, analyze users’ context, and prioritize learner’s motivation"). To me, though, these are  all varieties of xMOOC, and do not create the individual affordances of a decentralized cMOOC. You can't convert individual decision points and contexts into metrics without abstracting out essential elements of individual autonomy.

Web: [Direct Link] [This Post]

Testing the success of real-time online delivery channel adopted by higher education institutions in the United Arab Emirates during the Covid-19 pandemic
Iffat Sabir Chaudhry, Rene Paquibut, AbuReza Islam, Habib Chabchoub, International Journal of Educational Technology in Higher Education, 2021/09/09

It's interesting to read this post in the context of the article on measurement I just cited. This post "assessed the success of the e-learning system adopted by the higher education institutions in UAE during the pandemic" using an "information systems success model, including the measures of: information quality, system quality, system importance, system prior-use, perceived usefulness, user satisfaction and future-use intent." Of these, "information quality was found to have the strongest direct and total indirect effect" and so "education institutions, therefore, must prioritize the quality of information and education imparted to the students." Now I would find this conclusion unintuitive (though I suspect many educators would not) because I think that things like access, relevance and learning design are more important. But of course I can't simply depend on my feelings here; I would need to show that the study was constructed in such a way that access, relevance and learning design were factored out, and so not measured (which is in fact what I think has happened).

Web: [Direct Link] [This Post]

What is experiential learning and how does it work?
FutureLearn, 2021/09/09

As the subhead says, "we explore Kolb’s theory of experiential learning, discuss its cyclical nature, and think about how to apply experiential learning in the classroom and beyond." This is "the idea that experiences are generated through our ongoing interactions and engagement with the world around us, and learning is an inevitable product of experience." Kolb in particular proposed a cycle consisting of active experimentation, concrete experience, reflective observation and abstract conceptualization. It is relevant (to me) that three of these four elements are cognitive, not experiential (that is, they are elements of planning, reflecting and theorizing) and that they are all intentional (that is, we make a specific effort to do them). My own take on experiential learning is that it is much less formal and intentional than described here. You can learn a ton by simply aimlessly exploring and taking it all in. But you won't necessarily learn specific and intended things; for that you need some sort of process, and Kolb's is as good as any, so long as we keep in mind that planning, reflecting and theorizing are things we bring to the experience, and not part of the experience itself.

Web: [Direct Link] [This Post]

Goodhart’s Law and Why Measurement is Hard
David Manheim, Ribbonfarm, 2021/09/09

Stella Lee just linked to this article from a couple years ago on LinkedIn and I wanted to pass it along. Measurement is really tricky, and here's why: "Measurement replaces intuition, which is often fallible. It replaces trust, which is often misplaced. It finesses complexity, which is frequently irreducible." But despite this - or perhaps because of this - measurement does not answer all questions. This is especially clear in the case of complexity. Converting measures to metrics (i.e., subject to variables we can change) reduce dimensionality, obscure causal relationships, and sometimes even change the systems they are created to measure. These are all stones in the road of evidence-based policy, and are reasons why we should treat claims that something is 'evidence-based' with caution (note: I am not recommending we just go with intuition as an alternative).

Web: [Direct Link] [This Post]