One of the sample datafiles that comes with MINITAB is GAGELIN.MTW. The dataset is reprinted with permission from the Measurement Systems Analysis Reference Manual (Chrysler, Ford, General Motors Supplier Quality Requirements Task Force). Five parts were selected by a plant foreman that covered the expected range of measurements and each part was randomly measured 12 times by an operator. The data are given below.

What graphical device would you use to show the variability in measurements by the operator relative to the reference value for each part that is given in the adjacent column? Be specific. Would a histogram of the measurements show anything of value? Explain.

One of the sample datafiles that comes with MINITAB, MASSCOLL.MTW in the STUDNT12 subdirectory of datafiles, contains various statistics for the 56 four-year colleges and universities in Massachusetts in 1995. Consider the variable Acceptance Rate, with the following data:

If you wanted to show the distribution of this variable over the 56 schools and spotlight extreme observations, what graphical technique would you employ? Produce the graph and comment. Over the years there have been suspicions (not necessarily for Massachusetts) that self-reporting of certain statistics such as average SAT score has resulted in exaggerated numbers being reported.

(a) Recognizing that there are many different variables in this datafile, how would you check to determine if any of the numbers looked out of line?

 (b) Which pair of variables would you expect to have the strongest linear relationship? Check your answer by constructing all possible scatter plots. (This can be done in MINITAB by using the command MATR followed by the columns for which the set of scatter plots is to be constructed.