Correlations
A correlation compares the rate at which some property or characteristic
shows up in two different groups.
Wealthy voters voted for the Republication candidate in the recent
election more than nonwealthy voters.
This can be written as: The percentage of wealthy people who
voted for the Republican candidate in the recent election is greater than
the percentage of nonwealthy people who voted for the Republican
candidate.
More precisely it can be written in two different passages:
1. X percent of the wealthy people voted for the Republican
candidate.
2. Fewer than X percent of the nonwealthy people voted for
the Republican candidate.
The properties are the same (voted for the Republican candidate)
and the populations are different (wealthy voters and nonwealthy
voters).
CR:
A
is correlated with
B
in population
P
= Within population
P
,
the percentage of
A
s who are
B
is greater than the percentage of
non
A
s who are
B
.
A
= the property that picks out one of the subpopulations (being
wealthy)
B
=the property in question (voting for the Republican candidate)
P
=overall population (voters)
Standard form for correlation statements
:
A
is correlated with
B
in
population
P.
(p. 260)
1
This preview has intentionally blurred sections. Sign up to view the full version.
View Full DocumentIn the example this would be written as being wealthy is correlated
with voting for the Republican candidate in the recent election.
This is an example of a positive correlation: 2 factors (being
wealthy and voting for the Republican candidate) are positively
correlated in population
P
(voters) when the percentage of
A
s (wealthy
people) who are
B
(voted for the Republican candidate) is GREATER
THAN the percentage of non
A
s who are
B.
A negative correlation would be when the percentage of
This is the end of the preview.
Sign up
to
access the rest of the document.
 Spring '10
 Dr.Mc.
 Democracy, Household income in the United States, Correlation and dependence

Click to edit the document details