b.
What does the scatterplot suggest about the relation between these two variables?

c.
Would it be appropriate to calculate a Pearson correlation coefficient? Explain your
answer.

d.
Construct a second scatterplot, but this time add a participant who scored 1 on
externalizing behaviors and 45 on anxiety. Would you expect the correlation coefficient to
be positive or negative now? Small in magnitude or large in magnitude?
12
10
8
6
4
2
0
45
40
35
30
25
20
Externalizing Behaviours
Anxiety
Scatterplot of Anxiety vs Externalizing Behaviours
e.
The Pearson correlation coefficient for the first set of data is 0.65; for the second set of

data it is 0.12. Explain why the correlation changed so much with the addition of just one
participant.

15.36
Direction of a correlation:
For each of the following pairs of variables, would you expect
a positive correlation or a negative correlation between the two variables? Explain your answer.
a.
How hard the rain is falling and your commuting time

b.
How often you say no to dessert and your body fat

c.
The amount of wine you consume with dinner and your alertness after dinner
Mental health and partial correlation:
A study by Nolan and colleagues (2003)
examined the relation be-tween externalizing behaviors (acting out) and anxiety in adolescents.

Depression has been shown to relate to both of these variables. What role might depression play
in the observed positive relation between these variables? The correlation matrix below displays
the Pearson correlation coefficients, as calculated by computer software, for each pair of the
variables of interest: depression, externalizing, and anxiety. The Pearson correlation coefficients
for each pair of variables are at the intersection in the chart of the two variables. For example, the
correlation coefficient for the association between depression (top row) and externalizing
(second column of correlations) is 0.635, a very strong positive correlation.