Since I wrote about the 3 P's of good feedback a couple of weeks ago I've had several conversations with colleagues which seem to need some amplification.
Not much to add to what I said previously - I suspect that for students prompt means 21 seconds or 21 minutes after submission, not 21 days - except that electronic submission of work has changed the game and is probably responsible for some of the dissatisfaction we face. When students had to trek across campus to a printer, print their work, then track back to a departmental office to stand in a queue to hand it in, 21 days seemed like a fair interval to wait for a response. But that's not how it is any more. Clicking a Submit button online generates an expectation of a much quicker interaction. For anything other than simple automatically marked MCQs, this just isn't going to happen in the interval (seconds or minutes) that students want. I don't know what we do about this, other than to get better at explicit telling students what to expect "After submission you will receive your marks and feedback within 24 hours / one week / whenever."
Group feedback does not satisfy the feedback beast. However, it could buy us some time, if it's sufficiently prompt. Group feedback on essays (good points and common errors) within 24 hours with detailed personal feedback to follow in 7 days? Feedback on exams is another huge problem that group feedback won't solve. The examination process used to take a few days, or maybe even hours for a small class. Now it takes weeks. This is partly because of student numbers, but mostly because of the huge backend bureaucracy we have built. Students don't understand this, it further distances us from them. And yet I can't see the bureaucracy going away, so the only inadequate solution I have is to better communicate with students why it takes us so long.
This is the idea that has generated most heat in discussions. "If we don't tell students what's wrong with their work how can they improve?". I was raised in an academic era when my work was bluntly, sometime brutally, criticized (when needed). Crap was frequently scrawled in the margins, and it's hard for me to break out of that mould. It was tough at the time, but you could argue that it made me resilient (or maybe I was already resilient, which enabled me to survive). Either way, it doesn't work any more. If feedback is not positive, non-resilient students switch off and disengage. But that doesn't mean you can't tell students what's wrong with their work. Rather than writing "You didn't include any diagrams", say "If you include some diagrams in your next essay you will get better marks". One consequence of this is that the classic sh*t sandwich feedback formula is now beyond the pale. We're in the unbearably upbeat Have A Nice Day era. We need to adapt. Feedback is not peer review. Feedback is not performance monitoring, it is mentoring. Once again we have fallen down the crack between feedback and assessment.