Direct Teacher/Student interactions in the digital classroom

Between Clark and Warnock’s advocacy for the digital classroom and my own experiences, counting iLearn, Elluminate, etc., I must say that I’m totally convinced of the utilitarian benefits and think we should all switch immediately. There’s just one problem: that last clause was sarcastic. Here, let’s try it again: “I must say that I’m TOTALLY convinced of the utilitarian benefits, and think we should ALL switch IMMEDIATELY.” Catch the sarcasm that time? It’s ok if you didn’t; you’re not alone.

Embodied in that sarcasm is my overall issue with the digital classroom; while it offers many obvious benefits in both education and cost, it also has significant problems in terms of direct human interaction. First and foremost is the communication problem inherent in text-only conversation. Many studies have purported to tell how much of our communication occurs through non-verbal channels. One which said that only 7% of the message is carried by words is exaggerated by most who reference it, but even follow-up studies with more robust methodologies still state that nonverbal cues are somewhere between 20% and 80% of the message when we talk to each other.

While that’s a pretty big range, it is still helpful in that it’s very clearly not zero. The above sarcasm demonstration is just one small example in a very big problem. Even the perennially sarcastic RFC guidelines warn that sarcasm in text form “doesn’t travel well.” It’s taken us nearly 50 years of AI research to teach a computer to understand sarcasm sometimes, and if we can’t even expect a computer to always understand the intent of a message in text, how can we expect students to do better?

This simple problem has knock-on effects: studies have shown that if the primary mode of communication is text-only, authority figures appear more intimidating. For teachers and students alike, dealing with another human through text alone dehumanizes them. With students already in a position where they must appeal to an authority figure – say they’re behind on homework, or not understanding a key concept – might this additional hurdle of having to do so through a digital-only medium with all the potential for misunderstanding that brings be the straw the breaks the student’s back?

One final example: a longitudinal study once showed that after you normalize everything else about students, including their socioeconomic class, race, family history, literacy, grades in high school, and so on, after correcting for all those factors, the biggest single difference between college freshmen who graduated and college freshmen who dropped out was a single interaction with a teacher, usually outside of class. It could be office hours, or a brief chat before or after class, or even a chance meeting on campus between classes. If we remove the teacher and the student from sharing a campus – or even sharing a zip code – how will they have these interactions that save college careers? And if they don’t have them, is the digital classroom with all the benefits and savings worth all the students who might have had a degree if we hadn’t economized?

Hypertextual Analysis

In the video “The Machine is Us/ing Us,” Professort Wesch demonstrates many aspects of the power of hypertext. Its flexibility, its ability to separate form from content, its innumerable effects on things we take for granted, like privacy and free speech. The idea he seems to be getting at by emphasizing the ability hypertext has to link from what one reads to what the person writing it was reading, we are teaching “the Machine” – which is us – about the connections between things. When we link from one document to another, we are saying there is a connection between the two. When we tag a photo on Flickr or a video on YouTube, we are saying that this bit of media is related somehow to all those other bits of media with that tag, if only by sharing a subject.

What is unclear is whether the meaning inherent in all of the linking hypertext contains is being interpreted correctly. An implied link is there, but like all implication, the interpretation of the reader shapes meaning. When I linked to the EFF above, is there an implied assumption that I agree with their mission? What about the “Echelon” Wikipedia page? Do I agree with the “neutral” opinion therein displayed? Have I even read that page? The reader can not know. Maybe I just googled “internet privacy” and copied the first result that looked informative. Maybe not.

The point is that the reader must guess what I mean by any link or tag I place in hypertext. And, even if they guess correctly, there are still uncertainties. For example, assuming I actually read the “Echelon” page I linked, one must continue to remember that Wikipedia is the free encyclopedia that anyone can edit. Maybe when I clicked on that link, there was some spurious misinformation placed there by an anonymous troll. Maybe there wasn’t, but some was added before the reader of this post clicked it. Without reading the edit history, the reader can only hope. Even with the edit history, it’s speculative whether the reader and I actually read the same source.

Even worse, maybe the page I linked isn’t even there anymore. Dead links are increasingly common the further back one reads. While the Wayback Machine and Google Cache help, they only ameliorate the problem; they do not solve it.

One last problem with hypertextual analysis is simply that not everything has been digitized just yet. If I read something new and exciting on Slashdot, I can give the reader a link with reasonable certainty. If I read something interesting in an arbitrary relatively obscure book from the 1980s, the best I can give the reader is a link to the Amazon.com page and hope they can track down a copy to read themselves. Which is a shame, because that is actually a really good book.