Cost Curves: Don't "Let Me Google That for You"

It's amusing reading students' responses to take home exams. Some would clearly benefit from a even the most meager level of effort, while others show a clear and sometimes deep understanding of the material. Then, there are the kids who Google shit.
Of course, when they do this, they're rarely clever enough to put information they take from Wikipedia or Quizlet into their own words, and it's not too hard to catch them by selecting the plagiarized text and search for it using the context menu in Chrome (or Firefox).
But, I digress. In my police work, I stumbled across the Wikipedia page for "Cost Curves." Near the bottom there is a heading labeled "Cost Curves in Reality." Under this heading, the entry notes,

The U-shaped cost curves have no basis in fact. In a survey by Wilford J. Eiteman and Glenn E. Guthrie in 1952 managers of 334 companies were shown a number of different cost curves, and asked to specify which one best represented the company’s cost curve. 95% of managers responding to the survey reported cost curves with constant or falling costs.
Alan Blinder, former vice president of the American Economics Association, conducted the same type of survey in 1998, which involved 200 US firms in a sample that should be representative of the US economy at large. He found that about 40% of firms reported falling variable or marginal cost, and 48.4% reported constant marginal/variable cost.

The question is: Should I be surprised at these findings? It doesn't seem that these empirical findings suggest anything abnormal about the actual costs of production.
Why? First, it would seem to me that managers most likely are familiar with the range of production below and including the current range of output they are producing. If the industry is anything other than perfectly competitive, the current range of output will be on the decreasing side of average costs. Second, technology reduces costs over time, and therefore as firms expand they are also improving technology, and thus reducing their costs. None of this disproves the existence of upward sloping average costs at some level of output way out past the firm's current production horizon.
Now, what would be interesting would be to put this on an upcoming take home exam to see how many students Google it, and hit to this post. Will they have the gall to plagiarize their own professor?