Undergrads avoid OOP

[Joakim G.]

Interesting thread on comp.lang.java.programmer:

On Fri, 7 Feb 2003 15:33:08 -0000, "Graham Perkins"
<gperkins@REDACTED> wrote:

>>I've been teaching Computing undergraduates for twenty years.
>>I can assure you that their early abilities at understanding tasks
>>in terms of sub-tasks are picked up extremely quickly, perhaps
>>partially built-in to their cognitive abilities, and easy to extend.
>>
>>Their ability to understand processes in terms of collaborations
>>of behavioural units is much more primitive.  Many of them never
>>get it.  Almost *all** undergrad students avoid OOP in their
>>final year projects.
>  
>

Now that, is interesting.


>>>>> > You are claiming OOP has an advantage over other styles of
>>>>> > programming because it is supposedly easier to learn...
>>>      
>>>
>>
>  
>
>>>> I didn't intend to make that claim.  I claim that object oriented modeling
>>>> draws upon natural cognitive strengths evolved for predicting social
>>>> interactions.
>>    
>>
>>
>>You are trying to make an intellectual argument for that claim.
>>I'm telling you the hard facts: the students simply cannot do it.
>  
>

Suddenly I want to study this in depth.  

Have you confirmed this with instructors at other schools?  Heck, have
you confirmed it with other instructors at your same school?  Is there
something about the way your institution teaches OOP that makes it
seem harder than ordinary Java programming?  If this is a real
phenomenon, how is it students graduate and then are "forced" to do
OOP in so many situations?  

What *is** their favorite choice of tool for final year projects?

Joshua Stern