"tutorny" <firstname.lastname@example.org> wrote in message news:email@example.com... > >> Do you really think there is a 97% chance that >> there will be a 30% drop out rate? Suppose I asked >> you the probability that 100 of the students would >> drop out. How would you calculate that? Do you think >> that is a higher probability than only 30 students >> dropping out? > > I am not entirely sure that I follow. 30 students out of 110 is 27%. > So if a 20% rate is to be expected, isn't it feasible that a 27% rate > is 96% likely? > > If I made the mistake there, it would be in making p = 0.20. So, I > think what you are saying is that p should be = 0.80. But doing the > calculations that way results in a nonsensical answer..
No, there is a 50% probability of there being a drop-out rate of greater than 20%. At the extreme of a greater than 27% drop-out rate there is likely to be a very small drop-out rate ie in the tail of the distribution.