I have a student this term -- let's call her Charlene -- who, early in the term, appeared in my office hours requesting help with the first assignment. I am strongly inclined to help people whenever they ask, and anyway it's always encouraging to see students sufficiently engaged to come to office hours. Naturally I helped her with the assignment, which was to create a small web site consisting of three pages. She seemed to get it, and I was pleased. She didn't make any changes after our session during office hours, but I chalked that up to busyness and an impending deadline.

The next assignment was to produce a revised version of the same web site, incorporating new techniques we'd covered in class in the meantime. She showed up at office hours again, and again I helped her. Once again she didn't make any more changes after the session.

The third assignment of the term is notably more challenging than the previous two -- it involves changing the visual appearance of a web page without altering its HTML. The students have to use CSS instead. Anyway, we had a work day in class for this one. During the work day I floated around the room fielding questions from students working on their code. Charlene kept flagging me down to ask very basic questions which she could have resolved quickly by trying out her ideas. "Well, try it out," I told her. "But I make mistakes!" she replied in consternation.

At that point it became clear that Charlene had learned fairly little during the previous two assignments. She came to office hours not because she was extra interested in the material, but because she was afraid of making mistakes. Looking back, I realized that both of her first two assignments had been completed entirely in office hours. She hadn't spent any time at all working independently. She had wanted me to tell her exactly what to do. And I did. And she did it.

In terms of process, a great deal of web design consists of trial and error. Maybe you make an initial guess that a menu needs to be 150 pixels wide, but it turns out to be too narrow, so you increase it to 200 pixels, and it's too wide, and so on until you arrive at the perfect width (say, 183 pixels). But Charlene is afraid of making mistakes, and so she's contrived to avoid that process as much as possible. With my well-intentioned cooperation.

There in class that day when she said "But I make mistakes!" I paused for a moment fumbling for something to say. Then I replied: "Good! You need to make mistakes. You learn more from screwing up and then fixing it than you do from getting it right the first time." And I left her to it (there were other students with questions). Judging by the outcome of her third assignment, Charlene persuaded one of her classmates to take my place as hand-holder -- there were pronounced similarities between Charlene's finished site and one of those belonging to another student in the class. This is not plagiarism or collusion; I had explicitly instructed the students to help one another that day.

I need to encourage Charlene to take risks, to allow herself to make mistakes so she can learn from them. But my entire course structure is working against me. In a technical class like this one, I adopt a "guided experimentation" approach to teaching, in which I demonstrate a technique and ask the students to follow along, then alter it to see what happens. But that doesn't work unless the students actually make alterations on their own initiative; if they just copy me, it's not likely to help.

I also actively encourage collaboration. Partly that's because I can't be everywhere at once, and partly because nothing teaches you something more thoroughly than having to teach it yourself. A student who has mastered something and then explains it to a colleague who hasn't gotten it yet will know it even better for doing so.

In Charlene's case, that means that she can turn to other students to support her. She's very good at making puppy-dog eyes, by the way. Worse, most of the rest of the term is dominated by a group project, so she's going to have formal arrangements with team mates to draw on.

I'm not sure how to proceed. My usual approach of helping more is not going to work, and indeed is counter-productive. The problem is basically behavioral: I need to persuade her to take risks, but she's risk averse. This may be the most thorny case I've faced to date.

Suggestions?