Quiz 1 for Monday 2 February 2009

Quiz 1 for Monday 2 February 2009 - Group Leaders for...

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: Group Leaders for Monday 2 February 2009: Ash Kevin D Beaghley Andrew J Conner Andrew D Dzotsi Kofikuma Adzew Gonzalez-Roglich M Hein Andrew M Joury Nicole Mazal Larbi Peter A Linden Stephan Mayor Jordan R McEdward Lari Ann ANYONE WHO IS NOT IN THE CLASSROOM AT THE START OF CLASS MUST WORK ALONE AND CANNOT JOIN ANY GROUPS ALREADY WORKING. Group Leaders: Your job today is to lead a short discussion of the paper sent to everyone last Friday. To that end, 1) please pick 2‐3‐4 other students with whom you will work today. Try to get a variety of students (not just people in your major alone) in order to get a variety of opinion and background. 2) keep a summary of the consensus findings of your group. This is to be handed in at the end of the discussion with everyone’s name on it. Everyone: In order to intelligently discuss the paper, you should start by reviewing the methods. So some things to look for concerning the two principal component analyses described in the paper: 1) How many variables were used? 2) How many observations were used? 3) How many PC axes were chosen by the researchers? 4) How did they decide on that number? Do you agree? 5) Are the methods used well explained? Can you tell whether they used PCA with correlation matrices, raw data or covariance matrices? 6) Do you believe that the tests of differences of the mean PC5 (PC7) scores for the purebred versus the hybrids are statistically defensible? Please comment on the overall design of the behavior part of the study with respect to the use of PCA to summarize the results of three distinct treatments. Can you think of any assumptions they are making about the relationships among the variables under the three different treatments? How about for the two groups, purebred and hybrid? Please comment specifically on the adequacy of the sample sizes to obtain good PC axes for interpretation and subsequent analyses. Any recommendations as to how to change the study? Finally, please point out any other good or poor aspects of the study with respect to the usefulness of the data to make the inferences drawn by the researchers. ...
View Full Document

Ask a homework question - tutors are online