Let’s focus on the sample of men with a high school education; investigations
of other groups and the di¤erences between them are given in the exercises at
the end of this chapter. An application of the one-samplettest yields
x¼ 1 : 38
sx¼ 1 : 53
SEðxÞ¼
1 : 53
ffiffiffiffiffiffiffiffi
476
p
¼ 0 : 07
t¼
1 : 38
0 : 07
¼ 19 : 71
It can be easily seen that the di¤erence between self-reported height and mea-
sured height is highly statistically significant (p< 0 :01: comparing 19.71 versus
the cutpoint of 2.58 for a large sample).
7.3 COMPARISON OF TWO MEANS
Perhaps one of the most common problems in statistical inference is a com-
parison of two population means using data from two independent samples; the
sample sizes may or may not be equal. In this type of problem, we have two
sets of continuous measurents, one of sizen 1 and one of sizen 2 , and we con-
sider the null hypothesis
H 0 :m 1 ¼m 2
expressing the equality of the two population means.
To perform a test of significance forH 0 , we proceed with the following steps:
- Decide whether a one-sided test, say
HA:m 2 >m 1
or a two-sided test,
HA:m 10 m 2
is appropriate.
- Choose a significance levela, a common choice being 0.05.
- Calculate thetstatistic,
COMPARISON OF TWO MEANS 253