EDMS 771 Multivariate Data Analysis
QUIZ 2 Spring 2008 Name:_________________________
NOTES: (A) Use precise technical terminology when appropriate: say “heterogeneity of variance”; don’t say “the variances are kind
of different from each other.” (B) The references in parentheses refer to pages in the text – useful for review – you may not look now!
(1) On the back of this sheet, sketch the Scree Plot for the PCA; be as accurate as possible but it is not
necessary to use a ruler or straightedge. (p 121)
0
0.5
1
1.5
2
2.5
3
3.5
4
4.5
1 2 3 4 5
(2) For the PCA, the total variance with respect to all the variables is: 5 (p 117-8)
(3) How many principal components would be needed to account for 80% of the variance: 2
(4) Which variable do you think is negatively correlated with all of the other variables? Infant mortality
(5) Explicitly, what does the value, .902, represent in the Component Matrix? (p 102)
Correlation of principal component with %People who Read variable
(6) If you want to compute component scores for each of the 107 nations, what transformation is required for
the variables being analyzed? (Class)
z-score (standard score)
(7) Explicitly, how was the value .602 in the Communalities for the Principal Axes FA computed? (Class)
Multiple R2 of %People Living in Cities = DV with other 4 variables as predictors
(8) What is the main reason for rotating factors in FA? (p 123)
Improve interpretability
(9) According to the textbook, the minimum value of a factor loading for practical significance is: .5
(p 128)
(10) The factor, not PC, loading for the variable, People Living in Cities,” is equal to:
.607 .779
(Class)
