SIGCSE 2013 Preview: Measuring attitudes in introductory computing

Brian Dorn and Allison Elliott Tew have been working on a new assessment instrument for measuring attitudes towards computing.  They published a paper at ICER 2012 on its development, and the new SIGCSE 2013 paper is on its initial uses.

In general, we have too few research measures in computing education research.  Allison’s dissertation work stands alone as the only validated language-independent measure of CS1.  Brian and Allison have been following a careful process of developing the Computing Attitudes Survey (CAS).  They’re developing their instrument based on a measure created for Physics. The Physics instrument has already been adapted for Chemistry and Biology, so the process of adaptation is well-defined.

What’s particularly cool about CAS is that it can be used as a pre-test/post-test.  What were the attitude effects of a particular intervention?  The SIGCSE 2013 paper describes use of CAS in a set of pre-test/post-test situations.

Here comes the remarkable part.  In the other fields, an introductory course actually leads to decreased interest in the field (more specifically, in attitudes less-like experts in the field).  But not in computer science!  The CAS indicates increased interest in the field after the first course.

Why is that?  I like the hypothesis that Brian and Allison suggest.  Students have some clue what physics, biology, and chemistry — but it’s probably significantly wrong about real practice, and real practice is more rigorous than they thought.  Students have almost no clue what computer science is. They probably have misconceptions, but they are not tightly held — we’ve found that high school students’ perceptions of what CS is can be changed pretty easily.  After a first CS course, students realize that it’s more interesting than they thought, so attitudes become more expert-like and positive.