> On May 13, 2018, at 4:36 PM, Stephen Foster <[email protected]> wrote:
> 
> Hi!
> 
> I was about to start building a tool for statically analyzing student code.  
> But I first want to ask if there's related work out there.
> 
> I'm interested in relatively simple stuff -- e.g. How many functions did the 
> student write?  How many expressions?  What's the average nesting depth of 
> expressions?  Etc.  Basically, I want to start quantifying things about 
> students' coding styles -- both to identify potential problems and also to 
> help students set goals (e.g. "Today, you wrote 2 functions.  Tomorrow, I 
> want you to try to write 3!").
> 
> Before I start writing this myself, is there anything that I should know 
> about?  Has someone already done exactly this?  Or is there some library for 
> static analysis or code metrics that I should be building upon?  I searched a 
> bit and couldn't find a lot.  Maybe I'm using the wrong keywords.


Non answer but somewhat related: 

I have written (Racket) scripts in the past when we used svn repos to submit 
code The basic answer is that 90% of the students wrote 90% of their code in 
the last 10% of the time available. Then they would say during public code 
reviews that they ran out of time and therefore couldn’t follow the (HtDP) 
design process. At which point I reply with “no you ran out of time, because 
you didn’t follow the design process.” 


-- 
You received this message because you are subscribed to the Google Groups 
"Racket Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to