In: Statistics and Probability
An instructor of a statistics course had students report data about themselves, including hand preference (left or right handed) and pulses rate (beats per minute). We will use these data to conduct a hypothesis test to answer the question "does mean pulse rate differ for left-handed students, as compared to right-handed students?"
A. What are the hypotheses for your test?
B. Do you have paired samples or independent samples?
A. The null and alternative hypothesis
where ,
and
are mean pulse rate for left handed and right handed people
respectively.
Note : Our research hypothesis is " does mean pulse rate differ
between left handed and right handed students." First we
state the null hypothesis , H0 , which is the hypothesis of no
difference. Thus in the null hypothesis , we say that there is no
difference in mean pulse rate (
) , and the alternative hypothesis is our research hypothesis . We
can see that the alternative hypothesis is non directional, as our
test to find out whether there is a difference.So the test is two
tailed . We are not interested to know whose pulse rate is more .
In that case the alternative hypothesis would have been
( mean pulse rate for left handed more than right
handed people) or
(mean pulse rate for left handed less than right handed
people) . And the test would have been one tailed.
B. The data is for two independent samples : pulse rate of left handed people and pulse rate of right handed people. Thus we have independent samples .
Note : Independent sample t test also known as two sample t test is used to conduct the test.
Note : Paired samples are paired observations on the same set of subjects under two circumstances . For example weight of a set of individuals before and after a special diet . Then we perform paired sample t test.