"The Social Services Safety Net Is Fraying" reported in March 2005 on the growing problem of waiting lists for mental retardation services in a certain state.8 The data consist of numbers of emergency cases and critical cases in a given month over a series of many months:
Students are to report which one of the conditions for a regression model is clearly not met; who has the best answer? Adam: The problem is that the sample pairs aren't independent. If there are a lot of cases on the waiting list for one month, then the next month would also have a lot of cases, because it would include leftovers from the month before Brittany: I don't know about the rest of the scatterplot, but those three points don't fall on a straight line, so the data aren't linear like they're supposed to be. Carlos: The sample size might not be large enough. That could be the problem because maybe the number of cases doesn't follow a normal distribution. Dominique: We don't have enough data values to see what the spread looks like but I think the spreads might not be uniform from one end of the line to the other.