The Washington Post reports this morning (apologies if this is behind a paywall) about how some universities are (finally?) moving from in-class lecture as the basis for their “large lecture” courses to the flipped or inverted classroom. Says the article:
Colleges are absorbing lessons from the online education boom, including the growth of massive open online courses, or MOOCs. And some professors are “flipping” their classrooms to provide more content to students online and less through standard lectures.
William E. “Brit” Kirwan, chancellor of the University System of Maryland, said the system hopes the redesigned courses save money and boost performance.
“The passive, large lecture method of instruction is dead,” Kirwan said. “It’s just that some institutions don’t know it yet. We do.”
This is nice to hear, but watch out for that phrase, “saves money and boosts performance”. Continuing on:
No one involved in these experiments claims that raising class size by itself leads to improvement. Colleges everywhere say the strength of the relationship between a student and a professor is crucial to learning.
But in an era of tight cost constraints, educators say it is equally crucial to set aside old thinking about course configurations and lecture methods and be willing to use computers and other technology in new ways to spur student engagement.
UMES, a historically black college in this small town in Somerset County, found that grades rose in chemistry, psychology, biology and visual arts even as more students were packed into each class.
Emphases above are added to highlight the two ideas in tension here. On the one hand, many large universities understand the plain fact that getting students personally invested in their education — what we commonly refer to as “engagement” — is an essential ingredient to student success. On the other hand, they want to pack as many rear-ends into the seats as possible and keep them there, because enrollment plus retention equals revenue. Neither of these concepts is wrong by itself, but sometimes I fear we adopt good pedagogical practices for all the wrong reasons.
I had a discussion recently with the Dean of the engineering school of a major university about peer instruction and the flipped classroom. The Dean was very enthusiastic about implementing these teaching platforms in the classroom. When I asked him why he was so enthusiastic, he said that these instructional methods provide “a higher level of throughput” in their courses — meaning that with the flipped class and peer instruction, they could have larger and larger course sizes and get students through the courses faster, ergo more students going through the program (and hence more revenue both from tuition and state appropriations).
Look, I’m sensitive to the job that administrators have and how difficult it can be to manage a large institution or a large department within an institution. But throughput? This is a term commonly reserved to discuss how fast bits can be pushed through a data network. But students are not bits. They are human beings, each with a unique backstory, a unique collection of conceptions and misconceptions they bring to class, and a unique combination of needs and skills when it comes to learning. The flipped classroom and peer instruction, along with many other newer pedagogies, are promising precisely because they allow instructors to connect in a deeper and more personal way with these students, even if it’s just being able to connect briefly with small groups in a large course or to let students participate in a large lecture with clickers.
When people use some kind of economy-of-scale argument to say that the flipped classroom is useful because it allows for “bigger and cheaper” courses, then the entire notion of the flipped classroom has itself become flipped. Yes, these methods can lead to higher levels of engagement in larger class sizes. But what ought to matter first and foremost is that the flipped classroom, peer instruction, and the like improve student learning — a claim with over 20 years of data in its favor [PDF]. Shouldn’t that be enough? Or am I just not being pragmatic enough?