Chapter IV

HYPOTHESIS TESTING FOR

PARAMETRIC TESTS


Fundamentals and Processes of Statistics
Hypothesis Testing
A statistical hypothesis is an assertion or conjecture
concerning one or more populations.
Null hypothesis: This refers to any hypothesis the
researcher wishes to test and is denoted by H
0
.
It is also known as no difference relationship hypothesis. It
implies neutrality and objectivity which must be present in
any research undertaking.
Alternative hypothesis : The rejection of H
0
leads to the
acceptance of an alternative hypothesis, denoted by H
1
. It
is the opposite of the null hypothesis. It specifies an
existence of a difference, and is therefore non-directional.
On the other hand, this hypothesis is also called a
predictive hypothesis. The predictive hypothesis specifies
that one group is better than the other, and is therefore
sometimes directional.
Type I and Type II Errors

Rejection of the null hypothesis when it is true is
called a type I error.
Acceptance of the null hypothesis when it is false is
called a type II error.

Possible Situations for Testing a Statistical Hypothesis
H
0
is true H
0
is false_______
Accept H
0
Correct decision Type II error
Reject H
0
Type I error Correct decision

Level of Significance
The degree of error that can be allowed for a
particular study is called the level of significance. It
is denoted by the Greek letter o. (It is the
maximum probability of committing a type I error).


Critical value and Critical region

The critical value separates the critical region
from the noncritical region.
The critical region or rejection region is the range
of values of the test value that indicates that there
is a significant difference and the null hypothesis
should be rejected.
The noncritical or nonrejection region is the
range of values of the test value that indicates
that the difference was probably due to chance
and that the null hypothesis should not be
rejected.




Level of Significance and Critical Region




Critical
Regions


Critical values

o
One-and Two-tailed Tests
A test of any statistical hypothesis, where the
alternative is one-sided, such as
H
0
: u = u
0
, H
0
: u = u
0
,
H
1
: u > u
0
, or perhaps H
1
: u < u
0
,
is called a one-tailed test.

A test of any statistical hypothesis where the
alternative is two-sided, such as
H
0
: u = u
0
,
H
1
: u u
0

is called a two-tailed test.
Steps in Hypothesis Testing
1. State the null hypothesis H
0
that u = u
0

2. Choose an appropriate alternative hypothesis
H
1
from one of the alternatives u < u
0
, u > u
0
, or
uu
0
3. Choose a significance level of size o.
4. Select the appropriate test statistic and
establish the critical region. (If the decision is to
be based on a p-value, it is not necessary to
state the critical region.)
5. Compute the value of the test statistic from the
sample data.
6. Decision: Reject H
0
if the test statistic has a
value in the critical region (or if the computed
p-value is less than or equal to the desired
significance level o); otherwise, do not reject
H
0
.
P- value

The p- value or observed significance level of
statistical test is the smallest value of o for
which H
0
can be rejected. It is the actual risk
of committing a type I error, if H
0
is rejected
based on the observed value of the test
statistic. The p-value measures the strength of
the evidence against H
0
.


P- value

Many researchers use a sliding scale to
classify their results.
If p < 0.01, H
0
is rejected. The results are
highly significant.
If 0.01 < p < 0.05, H
0
is rejected. The results are
statistically significant.
If 0.05 < p < 0.10, H
0
is not usually rejected.
The results are only tending toward statistical
significance.
If p > 0.10, H
0
is not rejected. The results are
not statistically significant.


Parametric Tests
These are tests that involve numerical data and
are based on assumptions about the parameters
of the population from which the sample is
drawn.
These are usually applied when the sample is
randomly drawn or normality can be assumed.
Commonly used parametric tests:
z-test t-test F-test
Product-moment correlation coefficient


One Sample Mean Test
Case 1
Test for a single mean when the population variance
is known
Test for a single mean when the population variance
is unknown and the sample is at least 30
Test-Statistic:

where:
= sample mean

0
= hypothesized value of the population mean
o = population standard deviation
n = sample size


n
x
z
/
0
o

=
x
One Sample Mean Test


Level of significance



Example 1
The average height of females in the freshman class of
a certain college has been 162.5 centimeters with a
standard deviation of 6.9 centimeters. Is there reason
to believe that there has been a change in the average
height if a random sample of 50 females in the present
freshman class has an average height of 165.2
centimeters? Use a 0.05 level of significance.

Test 0.01 0.05
One-tailed 2.33 1.645
Two-tailed 2.575 1.96
One-Sample Mean Test

Example 2
The average length of time for students to register for
summer classes at a certain college has been 50
minutes. A new registration procedure using modern
computing machines is being tried. If a random
sample of 35 students had an average registration time
of 42 minutes with a standard deviation of 11.9
minutes under the new system, test the hypothesis
that the population mean is now less than 50 minutes,
using a level of significance of 0.01. Assume the
population times to be normal.



One-Sample Mean Test

Example 3
It is claimed that an automobile is driven on the
average more than 20,000 kilometers per year. To test
this claim, random samples of 100 automobile owners
are asked to keep a record of the kilometers they
travel. Would you agree with this claim if the random
sample showed an average of 23,500 kilometers and a
standard deviation of 3,900 kilometers? Use a
significance level of 0.05.


One-Sample Mean Test
Case 2
Test for a single mean when the population variance is
unknown and the sample is less than 30

Test-Statistic:



where:
= sample mean

0
= hypothesized value of the population mean
s = sample standard deviation
n = sample size


1 ,
/
=

= n v
n s
x
t

x
One-Sample Mean Test
Example 4
An educational researcher found that the average
entrance score of the incoming freshmen in a famous
university is 84. A random sample of 24 students from
a public school was then selected and found out that
their average score in the entrance examination is 88
with a standard deviation of 16. Is there any evidence
to show that the sample students from public school
performed better than the rest in the entrance
examination using the 0.01 level of significance?


One-Sample Mean Test

Example 5
It is crucial that fire prevention sprinkler systems
become active quickly in case of fire. The times of
activation for a series of tests were (in seconds):
27, 41, 22, 27, 23, 35, 30, 33, 24, 27, 28, 22, 24.
The system was designed so that the true mean
activation time is at most 25 sec. Choosing the level of
significance to be 5%, does the sample support this
claim?


Two-Sample Mean Test
a. Test for the difference of means from independent samples
when the population variances are known
b. Test for the difference of means from independent samples
when the population variances are unknown and the samples
are at least 30
Test-Statistic


where:

1
= mean of the first sample

2
= mean of the second sample
d
0
= difference of the hypothesized values of two
population means
o
1
= standard deviation of the first sample
o
2
= standard deviation of the second sample
n
1
= size of the first sample
n
2
= size of the second sample


2
2
1
1
2
1
0
2 1 ) (
n n
d x x
z
o o
+

=
x
x
Two-Sample Mean Test

Example 6
All freshmen in a particular school were found to have a
variation in grades expressed as a standard deviation of
3. Two samples among these freshmen, made up of 20
and 50 students each, were found to have means of 88
and 85, respectively. Based on their grades, is the first
group really brighter than the second group at a 0.01
level of significance?


Two-Sample Mean Test

Example 7
A survey is made to determine whether college
graduates read at least four books per year more than the
non-graduates. A random sample of 66 college graduates
reports of reading an average of 9.3 books per year, with
a standard deviation of 4.6 books, while 78 non-
graduates report an average of 4.9 books per year with
standard deviation of 3.1 books. Test the hypothesis at
the 0.05 level of significance.


T-test for 2 Independent Samples

Test for the difference of means from independent samples when
the population variances are unknown and the samples are less
than 30
Test-Statistic:

where:

1
= mean of the first sample n
1
= size of the first sample

2
= mean of the second sample n
2
= size of the second sample
d
0
= difference of the hypothesized values of two population means
SS
1
= sum of squares of the first sample
=

SS
2
= sum of squares of the second sample
=


|
|
.
|

\
|
+
|
.
|

\
|
+
+

=
2 1
2 1
0
2 1
1 1
2
) (
n n n n
SS SS
d x x
t
x
x
( )
1
2
1
2
1
n
x
x

( )
2
2
2
2
2
n
x
x

T-test for 2 Independent Samples
Example 8
To find out whether a new serum is effective will arrest
leukemia, 9 mice, all with an advanced stage of the
disease, are selected. Five mice receive the treatment and
4 do not. Survival times, in years, from the time the
experiment commenced are as follows:



At the 0.05 level of significance can the serum be said to
be effective? Assume the two distributions to be
normally distributed.


Treatment 2.1 5.3 1.4 4.6 0.9
No treatment 1.9 0.5 2.8 3.1
T-test for 2 Independent Samples
Example 9
The data below represent the number of hours of pain
relief provided by the two brands of headache tablets
administered to 24 tablets. These subjects were randomly
divided into two groups and each group was treated
with a different brand.



Test at 0.05 level of significance that there is no
difference between the average number of hours of pain
relief provided by the two brands of headache tablets.


Brand A 5 8 7 3 4 8 7 5 6 7 5 5
Brand B 3 2 1 4 5 6 7 3 5 4 2 3
T-test for Correlated Samples
Test-Statistic:




where:
d = difference of the two values
= mean difference of the two values
n = number of pairs of sample



d
( )
) 1 (
2
2

n n
n
d
d
d
t
T-test for Correlated Samples

A program for reducing the number of days missed by
food handlers in a certain restaurant chain was
conducted. The owners hypothesized that after the
program the workers would miss fewer days of work
due to illness. The table shows the number of days 10
workers missed per month before and after completing
the program. Is there enough evidence to support the
claim, at o = 0.5, that the food handlers missed fewer
days after the program?

Example 10

















T-test for Correlated Samples




Example 11
Ten subjects were given an aptitude test on a controversial
issue. Then they were shown a film favorable to the ten
subjects and the same attitude test was administered.
Make a directional test at the 0.05 level of significance.
Pretest 16 18 16 24 20 25 22 18 15 15
Posttest 20 20 24 28 20 30 23 24 19 15
ANALYSIS OF VARIANCE (ANOVA)



The F-test is the analysis of variance. It is used to
compare means of two or more independent
groups. One-way ANOVA is used when there is
only one variable involved. The two-way ANOVA
is used when two variables are involved: the
column and the row variables. The researcher is
interested to know if there are significant
differences between and among columns and
rows. This is also used in looking at the interaction
effect between the variables being analyzed.
Like the t-test, the F-test is also a parametric test
which has to meet some conditions, and the data to
be analyzed if they are normal are expressed in
interval or ratio data. This test is more efficient
than other tests of difference.





ANALYSIS OF VARIANCE



ANOVA Table







SST = sum of squares of x (sum of x)
2
/N
SSB = (sum of each x
i
)
2
/n
i
(sum of x)
2
/N
SSW = SST - SSB



Sources of
variation
Degrees of
freedom
Sum of
Squares
Mean
Square
F-value
computed
F-value
tabular
Between
Groups
k - 1 SSB SSB/df MSB/MSW
Within
Group
(N-1) -(k-1) SSW SSW/df
Total N - 1 SST
ANOVA
Example 12
A researcher wishes to try three different techniques
to lower the blood pressure of individuals diagnosed
with high blood pressure. The subjects are randomly
assigned to three groups; the first group takes
medication, the second group exercises, and the third
group follows a special diet. After four weeks, the
reduction in each persons blood pressure is recorded.
At o = 0.05, test the claim that there is no difference
among the means. The data are shown.
Medication Exercise Diet
10 6 5
12 8 9
9 3 12
15 0 8
13 2 4











ANOVA
Example 13
A researcher wishes to see whether there is a
difference in the weight gains of athletes following one
of three special diets. Athletes are randomly assigned
to three groups and placed on the diet for six weeks.
The weight gains (in pounds) are shown. At o = 0.05,
can the researcher conclude that there is a difference in
the diets?
Diet A Diet B Diet C
3 10 8
6 12 3
7 11 2
4 14 5
8
6