C 17
C 17
TOPIC
HYPOTHESIS TESTING
SUBTOPIC
Student Name
Date Started
Date Completed
Number of Exercises
Solutions
Hypothesis testing
This is actually a statement about the probability distribution of the population concerned.
Very often, we formulate a null hypothesis for the purpose of rejecting it. Thus, the null
hypothesis is the assumed parameter which we will compare with the sample result. This null
hypothesis is rejected if it is unlikely that the sample result is so much different from the
hypothesized value.
The hypothesis which differs and is opposite to a given null hypothesis is called an
alternative hypothesis. It can take the forms
Step 1
State the null hypothesis and the alternative hypothesis. The null and the alternative hypotheses determine if we
will have to perform a one-tailed test or two-tailed test.
The two-tailed test is used if the hypotheses are of the form:
Null hypothesis
:
Alternative hypothesis
:
The one-tailed test is used if the null hypothesis and the alternative hypothesis are of the form:
Null hypothesis
:
Alternative hypothesis
:
Step 2
Specify the level of significance of the test.
Step 3
Determine the critical value or values of the test statistic that will be used to assess the truth of the
nullhypothesis.
Step 4
Select the test statistic.
Step 5
Determine the actual value of the test statistic.
Step 6
Make the decision. However, to make the right decision, we must minimize errors or flaws in
the reasoning.
Two types of error can occur:
A Type I error occurs when we reject the null hypothesis when it is true. When we
specify that the significance level of the test is 5%, then the probability of rejecting
the null hypothesis when it is true is 0.05. In other words, the maximum probability
of a type I error occuring is 0.05. We write this as a = 0.05. Thus, the type I error is
also the significance level that we specify for the test.
In a two-tailed test, the null hypothesis is rejected for values of the test statistic which are
example
example
A market research company informs a prospective mini market entrepreneur that the average
income per household in the region is RM60 000 per annum. The household income is
assumed to be normally distributed with standard deviation of RM8 000, based on an early
study. For a random sample of 120 households, the mean household income is found to be
RM57 000. Test the null hypothesis that the population mean household income is RM60 000
at 5 percent significance level.
30) because of
example
A market research company informs a prospective mini market entrepreneur that the average
income per household in the region is RM60 000 per annum. The household income is
assumed to be normally distributed with standard deviation of RMS 000, based on an early
study. For a random sample of 120 households, the mean household income is found to be
RM57 000. Test the null hypothesis that the population mean household income is RM60 000
at 5 percent significance level.
USING HYPOTHESIS TESTING METHOD.
10
11
12
13
Example
example
The nails produced by a machine are assumed to have a mean of 5.0 cm. A random sample
of 80 nails gives a mean of 5.2 cm and a standard deviation of 1.3 cm. Is there evidence that
the population mean has increased
(a) at the 5 % level?
(b) at the 1 % level?
14
Example
If the sample size of 300 is used and the same sample mean ot 5.2 and the same sample
standard deviation of 1.3 are obtained, test the hypothesis using 5% significance level.
15
example
An average of 80 out of every 100 patients has survived a particular severe infection. A new
drug was introduced. The drug was administered to a random sample of 16 patients with the
infection and 15 recovered. Does this provide evidence that the new drug is effective at 5%
significance level ?
16
Example
Samsy Company claimed that at least 95% of the parts which the company supplied on
contract to car manufacturers conformed to specifications. A random sample of 500 parts was
tested, and 49 did not meet specifications. Can we accept the company's claim at a 5% level of
significance?
17
Various Exercise.
1.
18
2.
When cars arrive at a certain T-junction they turn either right or left. Part of a study of
road usage involved deciding between the following alternatives.
Cars arc equally likely to turn right or left.
Cars arc more likely to turn right than left.
(a) State suitable null and alternative hypotheses, involving a probability, for a
significance test.
(b) Out of a random sample of 40 cars, n turned right. Use a suitable approximation to
find the least value of n for which the null hypothesis will be rejected at the 2%
significance level.
(c) For the lest described in (b), calculate the probability of making a Type II error
when, in fact, 80% of all cars arriving at the junction turn right.
19
20