I used to watch Veritasium [1] videos (way back) that vividly illustrated that even when told the correct information, people would not change their world view but instead give answers consistent with their pre-teaching naive views.

Learning is complex. If all that a "teacher" is doing is lecturing, then a video might be a reasonable replacement.

Even when the content is straightforward, my observation is that engagement with videos are awful. Heck, engagement in a lecture is awful, but at least the (human) teacher in the room can reasonably be expected to do things to engage the audience and probe their understanding. Some of that might be possible in a video format, for 5x the effort; follow up with examples if you know good ones. I know in the MIT OCW I have seen recordings of discussion sections (very effective), and that should be the bottom of the barrel as far as engagement when "honestly trying to engage learners".

You would be better off just trying to do the work, then looking up the information needed. That kind of works, and I bet that's what happens in the flipped classroom model. (Maybe 20% of the class watched the video? Stats welcome.)

I'm really just trying to point out that learning is complex. There's supposed to be a national What Works Clearinghouse [2], but I had a hard time finding useful information there.

[1]: https://www.youtube.com/@veritasium [2]: https://ies.ed.gov/ncee/wwc/