In The Atlantic Monthly, James Somers argues that email can improve first-year composition by providing immediate feedback:

But of course professors don't train writers the way coaches train athletes. Instead they do it obliquely, one paper, one small barrage of comments at a time.

The whole process is rather clumsy. In a typical semester-long university English class there might be three opportunities to showcase your writing, each one a standard one-to-three-thousand-word expository essay, each one born, for the most part, in a few miserable halfday blasts.

Whatever you produce joins dozens of other efforts in a stack whose growing thickness doesn't exactly thrill your professor. But still he'll soldier on and read, and maybe re-read, and mark up and grade each one. Along the way he might leave short contextual notes in the margin; he might write one long critique at the end; he might do both; or he might do neither and let your grade do all the talking.

Trouble is, no matter how detailed and incisive the feedback, by the time it gets back to you it's already too late -- and, in a way, too early. Too late because your paper has already been written, and what you really needed help with was its composition, with the micromechanics of style, with all the small decisions that led you to say whatever it is you said. And too early because even if the professor's ex post pointers make every bit of sense, a whole month might go by before you next get to use them.

In contrast, there's email feedback:

it's deliberate practice: goal-directed, supervised. It's unfolding in smallish chunks in a series of tight feedback cycles. The conversations can be referenced, excerpted and combined; there is a clear trail of progress. "By the time we've done our half dozen email back-and-forths about their thesis, a lot of the time I can see direct evidence -- and they can see direct evidence -- that it's gotten better."

Alex Reid disagrees:

Writers don't need instant feedback from a mentor every time they get stuck. They mostly need to figure out how to work through their own problems.

In the comments, Alex provides more information, basically arguing that writing is complex and that, despite students' beliefs, there is no "ONE way to write." From this perspective giving instant feedback reinforces students' beliefs rather than helping them become independent writers.

The problem with this perspective is that it doesn't consider evidence from research on the role of feedback in learning.

Somers mentions, for instance, the research on expertise:

Ericsson argues that to become an expert at something you have to log about 10,000 hours of practice, but not just any practice -- a kind of practice that includes an "active search for methods to improve performance," immediate informative feedback, structure, supervision from an expert, and "close attention to every detail of performance 'each one done correctly, time and again, until excellence in every detail becomes a firmly ingrained habit.'"

Now, although I wouldn't expect that students in first-year composition will become experts in one semester (or by the time they graduate), the practice of "immediate informative feedback" still applies to improving competence (as does Alex's goal of having students engaging in "an intensive, regular writing practice"). Similar to Somers' points, I've written before about the importance of immediate feedback:

Although learning and instruction may meet conditions 2-4, seldom is immediate feedback given in composition classes. In one semester, students might write from three to six essays, depending on the instructor, which means that feedback on essays is given every two to three weeks. In addition, the feedback of peer reviews generally takes place hours after the last version, unless a student pulled an all-nighter for an 8:00 am class. In this case, most of the feedback will be seen through a haze. The feedback of instructors usually occurs days later after they have looked at all of them.

The importance of immediate feedback with cognitive tutors has been demonstrated in teaching LISP, algebra, and geometry. In their abstract, Anderson et al. write,

Early evaluations of these tutors usually but not always showed significant achievement gains. Best case evaluations showed that students could achieve at least the same level of proficiency as conventional instruction in one-third of the time.

Those "best case evaluations" are in the lab where there are no distractions, but even in real classrooms, Anderson and Schunn (pdf) have found achievement gains equal to one letter grade. Learning is directly due to time on task, that is, practice. (Of course, practicing the wrong tasks leads to mislearning.) Thus, providing immediate feedback helps to eliminate wasted time in trying to figure out how to do something, which in turn, decreases the time required to learn a particular activity.

Having students figure things out on their own wastes time not simply from the figuring out process but also from the mislearning that can just as easily occur and which must then be unlearned.

So, having students learn to "figure things out" sounds like a worthy goal, but it doesn't seem to be taught directly. Alex writes,

Over the years it seems that students increasingly come to me with a desire to be told what to do. I won't tell them what to do, but I will try to give them some insight into how they might figure out what to do. This isn't a matter of their playing guess what's on my mind. There isn't a right answer. Or more precisely, if there is a thing I want them to do it's for them to figure out what to do on their own. That's what I want them to learn, and if there is no shortcut to learning that, then the only way for students to learn this is for them to actual to do it.

Apparently, Alex uses an informal, implicit approach for those students who take the initiative to see Alex. However, although there are no shortcuts to learning, the wasted time incurred by implicit approaches can be eliminated through examples and feedback (see Learning with Examples and Learning with Examples cont'd).

To eliminate wasted time, ACT-R Theory (see, again, Anderson and Schunn) proposes the following:

  1. Learning occurs sequentially through three stages of declarative, procedural, and automatic knowledge.
  2. Acquiring expertise in any field requires extensive practice.
  3. Practice is made effective through
    • accurate diagnosis of the task/rules,
    • examples and understandable explanations for the task/rules, and
    • feedback based on the examples and explanations.
  4. Effective time on task is the most important factor in learning.
  5. Learning occurs best when re-iterated at intervals.

Note that it is effective time on task, not just time. One can practice bad habits of writing as well as good ones. In terms of learning to write, then,

  1. the first step in helping students to figure things out is figuring out how writers figure out what they need to do—that is, make explicit Alex's insight to students in classroom instruction. (The diagnosis of figuring out is unlikely to be easy.)
  2. Then, a sequence of examples of that "figuring out" need to be created, along with explanations.
  3. Students then iteratively practice "figuring out" and receive feedback connected to the examples and explanations.

On point 3, repetition in practice is important for learning as learning and forgetting are governed by power laws. That is, as noted in Learning: One step forward, then forgotten,

(1) with practice, learning increases quickly; (2) with a lack of practice, retention of learning drops off quickly; and (3) the effects of (1) and (2) interact in a way that multiplies each other rather than just adds up.

To avoid students' perception that "there is ONE way to write" and to promote transfer to novel situations, examples should be chosen that provide different aspects of "figuring out" (read also Learning by Remixing and Building Blocks and Learning).

One point I've never figured out is, Why is it not good for teachers to intervene heavily with students who are at a novice or intermediate level while coaches of different sports at the professional level consider it their job to do so?