We have spent much of the last year listening to reports that online learning is failing students.  The story goes something like this: Covid put an end to in-person education, schools used Zoom or Google Meet to move classroom instruction online, students did terribly, therefore online learning doesn’t work. The story is usually bolstered by anecdotes highlighting one or another ridiculous aspect of the system.   

Because the move to online learning was a solution made in desperation, often without much preparation, the fact that it has so often failed students comes as a surprise to no one.  And yet, for those of us who have lived our lives in online education, the failure is less the weakness of online learning and more suggestive of a systematic failure of the standard model of teaching. 

The argument I present here is involved, and one that could be elaborated into a much longer document, but out of consideration for readers, I will endeavor to be brief. 

Beginning in the early 1990s, with projects from the National Science Foundation and the Sloan Foundation, the availability of online courses rapidly increased. Online schools and programs at all levels of education became commonplace by 2000. During the late 1990s and early ’00s, a wide body of research was published showing that there was no discernable difference between online and in-person courses, at least as far as content learning went. Work in the late ’00s and early 2010s took this finding a step further, showing that when properly done, online learning could produce not just content knowledge equivalence but equal success in the social-emotional dimension as well. (For those interested, a detailed discussion of this topic can be found here.) 

Why, then, did online learning in the Covid era so badly miss the mark?  

One is tempted to blame it on the technology, or lack of familiarity with the technology. And while this may have been a factor, a lack of preparation and a lack of familiarity have always been with us.  

One might also blame the failure on the fact that the online option was being forced on students and instructors, rather than being something that they opted into. But necessity has long been a reason for online learning. Plenty of students and instructors who succeeded in the courses in the ’90s, ’00s, and ’10s found themselves there for reasons other than personal choice. 

Nor can one attribute failure to the fact that teachers were not in control. In the Zoom classroom, where teachers control the camera and the mic and set the rules for interaction, they have a good deal more control than they ever had in the in-person classroom. 

Two things were markedly different. First, in comparing the Zoom classroom to the traditional online course, what stands out most of all is the complete absence of student autonomy and low expectations of independent student work.  

The Zoom classroom and the typical in-person classroom embrace a model of command-and-control. As long as the teacher can broadcast information to the students and control their mics, the teacher is satisfied. This is very different from online learning, both asynchronous and synchronous models, where students are expected to have autonomy and independence and teachers are expected to facilitate rather than direct. In the Zoom classroom, this expectation for student autonomy has come last, resulting in a classroom that, while similar to in-person learning, is markedly more passive than what has historically been the case online. 

This is one way that things have been different in the Covid era. The second difference has been the lack of informal interaction between students during class. While the roles of teacher and student have been preserved, the rich dynamic interaction that occurs among students and between students and teacher has been lost.  

From this, we can conclude two things. For a teacher-driven class to succeed we need either 1) expectations of student autonomy and independence, or 2) active interaction between students.  Having a teacher front and center is not sufficient and may not even be necessary. Perhaps in the traditional in-person classroom students learn in spite of the standard model of teaching, and not because of it? 

Rather than rushing to bring students back into the old model of teacher command-and-control, now is the time for schools to fundamentally rethink how they conduct in-person education.  It should be possible to combine the traditional strengths of successful online learning, as well as student independence and autonomy, with the advantages of traditional classroom learning.  

What should be done? More attention should be paid to student autonomy and the student community, and how these elements are essential to effective learning. Such a school would operate less on a model of the teacher as someone who drives the bus with the students strapped to their seats, and more on a model of teacher as air-traffic control, helping to ensure that students are progressing along flight paths that will get them to their destinations. This is not to say that students all work independently of each other. Rather they would work together when it made sense to do so while having the autonomy to work independently when that made the most sense.  

In such a school, the amount of time spent physically in controlled environments could be reduced from the 35 hours typical in high school to the 15–20 typical in college, while the expectations of work being done by students could increase commensurately. Having students do more while being in class less would certainly increase learning. With class time reduced, the premium would be on teachers handling the live, unscripted moments of class, while using asynchronous technology like flipped classrooms to be efficient during moments when students were being subjected to one-way, lecture-style interactions. 

Once these steps are taken, it becomes possible to dissolve other structures of command-and-control that are more likely to serve as headwinds than as tailwinds. Dimensions of school-focused time and space could also be loosened. Students could move at their natural rates of learning, driven by interest, ambition, and ability. Doing so rather than marching in lockstep with everyone through the same experience of the same curriculum at the same time would surely deepen understanding. Competence-based testing where teamwork is possible and encouraged, just like students will experience in the world of work, would help to increase the likelihood that assessment is a predictor of future performance rather than past behavior.  

The possibility for meaningful improvement is at hand. Let us not take the failure of so many classes this past year to cast aspersions on online learning. Let us place blame where it properly lies: The standard model of teacher-direct classroom instruction is fundamentally broken. The past year has only served to show us just how broken it really is.