Feedback on MATH35001 exam Jan 2016 ----------------------------------- Overall the exam returned a pretty bimodal distribution. Many of you clearly "got it" and did extremely well; others seemed to struggle with even the most basic concepts and I wondered where you'd been during the lectures/examples classes... Re the specific questions: Q1: Mainly fine; some (though very few) students are (still!) confused by index notation... Practically everybody lost a point or two in the traction question for a wide range of reasons: wrong sign; traction not evaluated at the wall; and all kinds of other crazy things like adding the two components of the traction (vector!) into a single scalar...) Q2: Classical bookwork question. An amazing number of students just jotted down arguments in a random order (e.g. cancelling du/dx long before its zero-ness is established etc.). I tended to mark this generously, though I really shouldn't have -- this is cowboy maths! Many people lost points for genuinely sloppy/incomplete arguments. Q3: Lots of extremely woolly arguments were used to show (rather than to simply state, which was many people did essentially) that u = U f(\eta). You wouldn't believe the number of people who used an exp(\lambda \eta) ansatz to solve the non-constant coefficient ODE. I'll have to have a word with your first year lecturer -- that German bloke with a beard... Q4: Yet more woolly arguments when trying to justify the form of the solution and plenty of "cheating" when it came to showing the partial results stated in the question. I do read your alleged derivations from start to finish, you know... Q5: What happened here? I'd actually added an example to the lecture to show how easy it is to do such problems (given that the solution of the PDEs is already provided!). Most people did very badly on this and most also lost points in the bookwork part and in the question about the Reynolds number at the end.