1. The file ROOMS.SYD contains seven variables and 10 cases.  Each case represents an excavated room, and each variable represents the count of a pottery type within that room.  Focus your analysis on types A, B, and C only (although you are welcome to look at the other types if you so desire).

• Make scatter diagrams and calculate correlation coefficients for all combinations of these three variables.  Also create a scatterplot matrix for these three variables.
• Transform the data into percentages (with respect to the total of all seven types) and repeat the scatter diagrams and calculation of correlation coefficients.
• Discuss and interpret your results.

2. Analyze each of the following datasets: PLAY1.SYD, PLAY2.SYD, and PLAY3.SYD.  Each dataset contains two variables: PLAY1 contains imaginary data; PLAY2 contains counts of micro-gravette segments and butts by square in a Perigordian level at Abri Pataud; and PLAY3 contains counts of grooved rim and thickened rim Parthian bowls by provenience from a site near Hamadan.  Be sure to illustrate and discuss what you see going on in each dataset.  (The INFLUENCE option of the PLOT command might be especially useful to you in this regard; this is also available through the Options | Plot Options menu on the scatter plot dialog box.)
 

(Note: This exercise is based on one invented by Robert Whallon, and used by him in a similar course at the University of Michigan.)
 

Datasets for this exercise (right-click to download):

• Rooms.syd
• Play1.syd
• Play2.syd
• Play3.syd

Back