Communications of the ACM

The curriculum should inspire students to view CS as both accomplishment and intellectual discipline.

Mark Reitblatt

Some of the divergence might be explicable by ambiguities in the questions, particularly ones where students are more likely to read it one way, and professors are likely to read it in another. Take this question:

"Research in computer science often develops really important ideas"

There are at least 2 ways to read this: "Is it the case that most (or close to) research projects develop 'really important ideas'" vs. "Is it the case that there are many 'really important ideas' that emerge from CS research often". In other words, in the first the frequency of "often" is in comparison to the NUMBER of research projects, and in the second it's derived against the magnitude of the number of ideas.

Was the survey tested on a smaller population for ambiguity or confusion? Was this possibility of correlated ambiguity considered in the analysis?