Hypothesis testing for multiple parameters
•
Testing the difference among multiple proportions
•
Sampling distribution for multiple proportions (Chi-square)
•
Testing the difference among multiple means
•
Why a t-test is not used to compare multiple means?

EXAMPLE
A manufacturer of stereo amplifiers has three assembly lines.
We
want to test whether the three lines do differ with respect to the
number (proportion) of defectives produced
.
Independent samples
of 30 amplifiers each are selected from the output of the lines, and
the number of defectives is observed.
The data are as follows
:
Conduct a test of the hypothesis given above at the 5% significance
level.
TESTING THE DIFFERENCE AMONG MULTIPLE
PROPORTIONS

In this case,
independent
random samples are selected that
result in
binomially distributed random variables Y
1
, Y
2
, …, Y
k
where Y
i
is based on n
i
trials with success probability
on each
trial
The null and the alternative are as follows:
What are the null and the alternative hypothesis for the example?