Re: [vox] code sleuths needed - cheating suspected

[Peter Jay Salzman <p@dirac.MAPSorg>]

begin donw@examen.com <donw@examen.com> 
> On Wed, Nov 13, 2002 at 11:06:58AM -0800, Peter Jay Salzman wrote:
> > dear all,
> > 
> > i suspect a student has copied a homework assignment from another
> > student.  call them "student 1" and "student 2".
> >
> > [snip]
> > 
> > i won't mention the reasons why i think the two codes are similar.
> > instead, i'll let other people analyze it themselves.  i first suspected
> > copying because as i read through student 1's code, i kept saying to
> > myself "this is my own coding style".  then i remembered that i wrote
> > part of student 2's code who came to me for help with the program.  the
> > two of them are friends and work together.
> > 
> > i'm curious to know what people think.
> 
> Hmmm...while I'm hardly an expert, a few points come up:
>   - Student 1's initial code example is very poorly commented (almost no 
>     comments).  Comments in the code turned in by student 1 for the 
>     assignment are almost idential to those in student 2's code.
> 
>   - Variable names are identical.
>   
>   - Indent style is inconsistent; however, this looks like a first-year 
>     class, so that can be attributed to students playing around and finding
>     a coding style they are happy with.
> 
>   - The only thing that really bugs me is that the equation bits in the 
>     middle (inside of the for-loop) are set up in an identical fashion.
> 
> A conclusion?  Not really much of one; it could be a copy, but then again, 
> the similarities can be accounted for because the two likely work together 
> and help each other out.  In my CS classes, when I was stuck on a problem,
> I'd look at some sample code, and certain things (variable names, specific
> algorithms) would often end up the same in whatever I was writing, just 
> because they were in cache at the time.
> 
> I'd make a judgement call as a professor.  How many assignments have been
> turned in?  Is this code uncharacteristic of Student 1's work; e.g., does 
> he usually turn in broken code, but this one works?  Do you have any other 
> reasons to suspect cheating?
 
hi don,

yeah.  quite a few things, but i still want to hear what other people
have to say.   but i'll give one reason that people may not catch:

the line:

   int    N = (b - a) / h + 2;

is just wrong.  i wrote this line in a rush while 30 students were
asking for my attention.  the "+ 2" is just a hack.  it's wrong.  it's
there because i stupidly relied on integer division.   it works for
b=10, a=0 but doesn't work for, for example, b=10.5 a=0.

another thing that caught my eye:

   if ( (fp = fopen("plotpoints.dat", "w")) == NULL)

inconsistent spacing of the outer parenthesis shows up in both codes.

there's a lot of other things, but i'd like to hear other comments
before i give my list.

pete

-- 
Fingerprint: B9F1 6CF3 47C4 7CD8 D33E 70A9 A3B9 1945 67EA 951D
_______________________________________________
vox mailing list
vox@lists.lugod.org
http://lists.lugod.org/mailman/listinfo/vox