218 PART 5 Looking for Relationships with Correlation and Regression
on people and find that there really is an association between weight and BP, you
may want to quantify that relationship. Maybe you want to say that every extra
kilogram of weight tends to be associated with a certain amount of increased
BP. Even though you are testing an association, the reality is that you believe that
as people weigh more, it causes their BP to go up — not the other way around. So,
you would characterize weight as the independent variable (X), and BP as the
dependent variable (Y). The following sections take you through the steps of gath-
ering data, creating a scatter plot, and interpreting the results.
Gathering the data
Suppose that you recruit a sample of 20 adults from a particular clinical popula-
tion to participate in your study (see Chapter 6 for more on sampling). You weigh
them and measure their systolic BP (SBP) as a measure of their BP. Table 16-1
shows a sample of weight and SBP data from 20 participants. Weight is recorded
in kilograms (kg), and SBP is recorded in the strange-sounding units of millime-
ters of mercury (mmHg).
TABLE 16-1
Weight and Blood Pressure Data
Participant Study ID
Body Weight (kg)
SBP (mmHg)
1
74.4
109
2
85.1
114
3
78.3
94
4
77.2
109
5
63.8
104
6
77.9
132
7
78.9
127
8
60.9
98
9
75.6
126
10
74.5
126
11
82.2
116
12
99.8
121
13
78.0
111