Confidence Interval on the Difference Between Proportions (2 of 4)
Therefore the proportions who solve the problem are 0.65 for the
psychology majors (p
1) and 0.41 for the physics majors
(p
2). The goal of the experiment is to estimate the
difference between the proportion of psychology majors in the
population that can solve the problem (π
1)
and the proportion of physics majors in the population that can solve it (
π
2). The statistic
p
1 - p
2 is used as an estimate of π
1 -
π
2. In this
experiment, the estimated value of π
1 - π
2 is
0.65 - 0.41 = 0.24.
To compute the confidence interval, the
standard
error of
p
1- p
2 is needed. The standard error is:
The estimated standard error is:
= 0.067