 |
 |
I have been asked the following by a colleague: Imagine two students, in a class of size n, who submit an assignment and they (and they alone) make a particular spelling mistake or typographical error. How would one go about estimating the probability of this occurring by chance? Let us suppose that they have not one mistake in common, but two. Now what is the chance of this occurring by chance? I guess that one might need additional information such as how many words are in the document, and how likely it is that any given word may be mis-spelled. If so, let's assume that there are w words and that it's a really really unlikely (and therefore potentially quite damning) error. Any ideas? I would also be interested to learn of any work that has been carried out on the use of statistics to detect plagiarism. As usual, please reply to me, not the list. Many thanks in anticipation of any help. Michaela Cottee [log in to unmask] %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%