When conducting multivariate statistical analysis there are two approaches sample q-mode and variable r mode.  What is the difference and when would you use each?

After you graduate you are hired by an environmental testing firm.  Your job includes assuring that the testing equipment is functioning properly.   The firm has three portable pH, dissolved O2 meters for measuring streams.  It appears that the measurements may not be consistent between the instruments.  Given that there is a stream behind the lab how would you go about determining if one of the meters is giving inconsistent results?

When conducting a statistical test on the computer the p value indicates what? How does this compare to the traditional method of using a pre-selected level of significance?

When using cluster analysis what is the goal of the analysis?  Why is this different than other methods of analysis?  

Below is a diagram of a multivariate analysis which used discriminate function analysis to attempt to determine the source of sediments in streams in Kenya.  The data is based on chemical measurements from streams.  Questions: 1) Do you think the diagrams support the conclusion of the authors that Agricultural land is the main source of sediment load?  This is in part based on the grouping of the sources.  2) The results look very different in a and b what accounts for the difference?  Do you think that excluding the unpaved track data is warranted or a method getting the results desired?     

The Χ2 and K-S test are used to test what?  How do they differ from the t-test and F-test?

What is multiple regression and briefly explain the technique.

What is the effect of closed data on the regression analysis and what are some methods to correct for closure?

Kriging, Counting cells and Thiesson polygons are all methods of what?  Why are these methods preferred?

What are Type I and Type II errors in testing a statistical hypothesis?  If you limit the possibility of a Type I error what happens regarding a Type II error?