Sie sind auf Seite 1von 49

TESTS OF

HYPOTHESES
TESTS OF HYPOTHESES

A
statistical hypothesis is an assertion or
conjecture concerning one or more populations.
Two Types:
1. Null hypothesis() - a hypothesis of no
difference.
The hypothesis to be tested
Represents what the investigation doubts to be true
If rejected, is accepted

2. Alternative hypothesis () the operational


statement of the theory that the researcher
believes to be true and wishes to prove.
TESTS OF HYPOTHESES
ALTERNATIVE HYPOTHESIS are classified as:
NON-DIRECTIONAL asserts that one value is
different from another (or others).
An assertion that there is a significant difference between
two or more summary measures.
Makes use of the not equal to () sign
Also called two-sided hypothesis

DIRECTIONAL an assertion that one measure is


less than (or greater than) another measure of similar
measure.
Involves one of the order relations, less than (<) or
greater than (>).
One-sided hypothesis
Examples:

1.
The average annual income of all the families of
Mapayapa District is Php36,000 (=Php36,000).
The average annual income of all the families
of Mapayapa District is not Php36,000
(Php36,000).
2. There is no significant difference between the
average life of brand A light bubs and that of brand
B light bulbs (). There is a significant difference
between the average life of brand A light bubs and
that of brand B light bulbs ().
3. The proportion of Metro Manila college students who
prefer the taste of Coca Cola is 2/3 (p=2/3).
The proportion of Metro Manila college
students who prefer the taste of Coca Cola is less
than 2/3 (p < 2/3).
Examples:

4. he proportion of TV viewers who watch talk


shows from 9:00 to 10:00 in the evening is the
same on Wednesdays and Fridays (=)
he proportion of TV viewers
who watch talk shows from 9:00 to 10:00 in the
evening is greater on Wednesdays than on
Fridays ().
TWO TYPES OF ERRORS
Four possibilities exist in connection
with the acceptance or rejection of a
null hypothesis:
1. The null hypothesis is accepted when,
in fact, it is true.
2. The null hypothesis is accepted
when, in fact, it is false.
3. The null hypothesis is rejected
when, in fact, it is true.
4. The null hypothesis is rejected when, in
fact, it is false.
TWO TYPES OF ERRORS
Four possibilities:
TYPE
I ERROR
1. The null hypothesis is When the researcher
accepted when, in fact, rejects the when in fact
it is true. it is true (possibility 3).
2. The null hypothesis The probability is
is accepted when, in denoted by .
fact, it is false.
3. The null hypothesis
TYPE II ERROR
is rejected when, in When the researcher
fact, it is true. accepts the when in
4. The null hypothesis is fact it is false
rejected when, in fact, (possibility 2).
it is false. The probability is
denoted by .
TWO TYPES OF ERRORS

1.
The average annual income of all the families of
Mapayapa District is Php36,000 (=Php36,000).
If we reject , when, in fact, the actual
mean does not vary significantly from
Php36,000, then we are committing a
Type I error.
2. There is a significant difference between the
average life of brand A light bubs and that of
brand B light bulbs ().
If we accept , when, in fact, the opposite is
true, then we are committing a Type II error.
TWO TYPES OF ERRORS
Table 1.1
Consequences of Decisions in Testing
Decision Hypotheses
is true is false
Accept Correct decision Type II error
Reject Type I error Correct decision
Most commonly used values of are 0.01 and 0.05.
The lower the value of , the lesser is the probability of rejecting a
true and the higher is the probability of accepting a false .
More likely to commit Type II error with = 0.01 than
with = 0.05
But when = 0.05, the probability of committing Type II error
is greater than when = 0.01
LEVEL OF SIGNIFICANCE
There are no convenient ways of finding
whether or not the hypothesis that we accepts
is the true hypothesis. Thus, we usually cannot
tell whether we made the correct decision or
committed an error whenever we accept or
reject a hypothesis.
In nearly all cases, the most that we can do is
to assume some degree of confidence with our
decision by specifying the probability of
committing a type I error, , which is more
popularly known as the level of significance.
LEVEL OF SIGNIFICANCE
By using a level of significance, we can determine
a critical value which defines a region of
rejection (or critical region) and a region of
acceptance.
In effect this critical value serves as the basis for

either accepting or rejecting a hypothesis.


The area of the region of rejection is given by the

level of significance.
When = 0.05, the region of rejection will have an area of

0.05; thus the region of acceptance will have an area of 0.95.


Choice of a significance level is arbitrary
ONE-TAILED AND TWO-TAILED TESTS

ONE-TAILED a test of any


statistical hypothesis where is
directional.
Characterized by a region of rejection
which lies entirely in one end of the
distribution.
TWO-TAILED a test of any
statistical hypothesis where is non-
directional
Involves a critical region which is split
ONE-TAILED AND TWO-TAILED TESTS

is formulated in such a way that the greater than


If
(>) sign is appropriate for the equivalent
mathematical statement, the region of rejection is
placed entirely in the right tail of the distribution.
But if requires the application of the less than (<)

sign, then the region of rejection is placed entirely in


the left tail.
Both tails will contain equal areas serving as critical

regions when the mathematical formulation of


involves a not equal to () sign.
ONE-TAILED AND TWO-TAILED TESTS

The type of test (whether one-


tailed or two-tailed) along with
the significance level must be
considered in determining the
critical value used in
connection with the acceptance
of .
ONE-TAILED AND TWO-TAILED TESTS
ONE-TAILED AND TWO-TAILED TESTS
Example 1
A manufacturer of a certain brand of rice cereal claims
that the average saturated fat content does not
exceed 1.5 grams per serving. State the null and
alternative hypotheses to be used in testing this claim
and determine where the critical region is located.
The manufacturers claim should be rejected only if is
greater than 1.5 milligrams and should not be rejected if
is less than or equal to 1.5 milligrams. We test
: = 1.5,
: > 1.5.
Nonrejection of does not rule out values less than 1.5
milligrams. Since we have a one-tailed test, the greater
than symbol indicates that the critical region lies entirely in
the right tail of the distribution of our test statistic .
Example 2
A real estate agent claims that 60% of all private
residences being built today are 3-bedroom homes. To test
this claim, a large sample of new residences is inspected;
the proportion of these homes with 3 bedrooms is recorded
and used as the test statistic. State the null and alternative
hypotheses to be used in this test and determine the
location of the critical region.
If the test statistic were substantially higher or lower than p =

0.6, we would reject the agents claim. Hence, we should


make the hypothesis
: p = 0.6,
: p 0.6.
The alternative hypothesis implies a two-tailed test with the
critical region divided equally in both tails of the distribution
of , our test statistic.
EXERCISES
1. In a certain undergraduate research class, the
following problems were given to the students to
investigate. Formulate a null hypothesis and an
alternative hypothesis for each of the research
problems and then, tell whether a directional or a non-
directional test is applicable.
a. Do male students feel the same way about the long-term
engagement as female students do?

Malestudents feel the same way about the


long-term engagement as female students do.
: Male students do not feel the same way
about the long-term engagement as female
students do.
*NON-DIRECTIONAL
EXERCISES
b. Is fluoride toothpaste more effective in
fighting cavities than non-fluoride
toothpaste?
The fluoride toothpaste is as effective as non-fluoride
toothpaste in fighting cavities.
: The fluoride toothpaste is more effective than non-
fluoride toothpaste in fighting cavities.
*DIRECTIONAL

c. Does the white cell amount among leukemia


patients vary more erratically than that of
healthy
The individuals?
white cell amount among leukemia patients does
not vary than that of healthy individuals.
: The white cell amount among leukemia patients
varies more erratically than that of healthy individuals.
*DIRECTIONAL
EXERCISES
d. Does the teaching method used in a mathematics class
affect the overall performance of the students?
There is no significant difference between the teaching
method used in mathematics class and the overall
performance of the students.
: There is a significant difference between the teaching
method used in mathematics class and the overall
performance of the students.
e. Is passing the licensure examination for engineers
*NON-DIRECTIONAL
dependent on the type of school the prospective
engineer graduated
There is no from?
significant difference between passing the
licensure examination and the type of school the
prospective engineer graduated from.
: There is a significant difference between passing the
licensure examination and the type of school the
prospective engineer graduated from.
*NON-DIRECTIONAL
EXERCISES

2. Under the Philippine judicial system, an


accused person is presumed innocent until
proven guilty. Suppose we wish to test the
hypothesis that the accused is innocent against
the alternative hypothesis that the accused is
guilty . What type of error, if any, is committed
if the court
a. convicts the accused when, in fact, he is innocent?
b. convicts the accused when, in fact, he is guilty?
c. acquits the accused when, in fact, he is innocent?
d. acquits the accused when, in fact, he is guilty?
STEPS IN HYPOTHESIS
TESTING

1. Formulate the null hypothesis () and the alternative


hypothesis ().
These two statements should be formulated in such a
way that if one is true, the other is necessarily false.
2. Specify the level of significance ().
3. Choose the appropriate test statistic.
4. Establish the critical region.
5. Compute for the value of the statistical test.
6. Make a decision and, if possible, draw a conclusion.
If the computed value of the test statistic is within the
region of rejection, we reject and accept . If outside,
either we accept or reserve judgment.
TESTING A HYPOTHESIZED VALUE OF THE
MEAN

determine whether or not the


To
observed difference between the
computed value and the hypothesized
value is significant, we formulate the
following hypothesis:
: There is no significant difference between
the observed value and the hypothesized
value of the mean ().
: There is a significant difference between
the observed value and the hypothesized
value of the mean ().
TESTING A HYPOTHESIZED VALUE OF THE
MEAN

Since
the parameter is known, the z statistic is
employed as the test statistic. Consequently, the z
score corresponding to is
x - mo
z=
sx
where the denominator represents the
standard error of the mean (or the standard
deviation of the sampling distribution of the
mean) and is computeds by the formula:
sx =
n
Example 3
An electrical company claims that the
lives of the light bulbs it manufactures
are normally distributed with a mean of
1,000 hours and a standard deviation of
150 hours. What can you say about this
claim if a random sample of 100 bulbs
produced by this company has a mean
life of 980 hours? Use a 0.05 level of
significance.
Example 3 (Solution)
Hypothesis:

= The average life of all the bulbs produced by the firm is 1,000

hours (.
= The average life of all the bulbs produced by the firm is not

equal to 1,000 hours (.


Level of significance: =0.05, two-tailed test
Test statistic: z statistic
Critical region: Reject if or
Computation: x - mo 980 - 1, 000
z= = = -1.33
sx 150 / 100

Decision: The computed value of z is within the are of acceptance (-


1.33 >-1.96). Therefore, we may accept ; that is, we may say that
there is sufficient evidence to support the claim of the manufacturer
that the average life of the bulbs it produces is 1,000 hours.
TESTING A HYPOTHESIZED VALUE OF THE
MEAN

In
cases where a small sample size is used
and the population standard deviation, s, is
unknown, then the t-test will be used:
x - mo
t= ; df = n - 1
sx
where the denominator represents the
standard error of the mean (or the standard
deviation of the sampling distribution of the
mean) and is computeds by the formula:
sx =
n
Example 4
A random sample of 100 recorded
deaths in the United States during the
past year showed an average life span of
71.8 years, with a standard deviation of
8.9 years. Does this seem to indicate
that the average life span today is
greater than 70 years? Use a 0.05 level
of significance.
Example 4 (Solution)
Hypothesis:


= The average life span today is 70 years (.
= The average life span today is greater than 70 years (.
Level of significance: =0.05, one-tailed test
Test statistic: z statistic
Critical region: Reject if
Computation:
x - mo 71.8-70
z= = = 2.02
s 8.9/ 100
x
Decision: The computed value of z is within the are of
rejection (since 2.02>1.645). Therefore, we must reject ;
that is, we may say that the average life span today is
greater than 70 years.
Example 5
A certain brand of laundry soap is
advertised to have a net weight of 500
grams. If the net weights of a random
sample of 10 boxes are 495, 503, 507,
498, 490, 505, 510, 502, 493, and 506
grams, can it be concluded that the
average net weight of boxes is less than
the advertised amount? Use 0.01 level of
significance.
Example 5 (Solution)
Hypothesis:


= The average net weight of laundry soap is 500 grams
(.
= The average net weight of laundry soap is less than
500 grams (.
Level of significance: =0.01, one-tailed test
Test statistic: t statistic
Critical region: Reject if the computed t, with df=10-1.
10

xi
i =1
Computation: The sample mean
x =
nis
= 500.9 grams
and the sample standard deviation is s=6.61 . Thus,
x - mo 500.9-500
t = = = 0.4306
s 6.61/ 10
x
Example 5 (Solution)
Decision: The computed value of t is
within the area of acceptance (since
0.4306>2.2821). Therefore, we must
accept ; that is, we may say that the
average net weight of the laundry soap is
really 500 grams.

When
we fail to reject the , we never say that
we accept the null hypothesis because we have
not proven that the is true.
The is assumed to be true at the beginning of
the hypothesis-testing procedure.
Failure to attain evidence to reject the in favor
of the does not equate to proove that the is
true.
It merely means that we did not have enough
evidence to reject the and thereby, failed to
reject it.
TESTING THE DIFFERENCE BETWEEN TWO
MEANS

There are 2 types of commonly used test


of significant difference between two
samples:
INDEPENDENT TEST used when the 2
sample means are taken from separate
groups of respondents or populations.
CORRELATED TEST applied to matched
pairs of correlated samples taken from one
population.
Example 6

An
instructor wishes to determine which of two
methods of teaching method A or method B is
more effective in teaching a certain concept in
statistics. In a class of 36 students, he used method A
and in another class where there are 40 students, he
used method B. He gave the two classes for the same
examination and obtained the following results:
Method A Method B
Mean 78 70
Standard 4 6
deviation
Is he correct in assuming that Method A is more
effective than Method B? Use .
Example 6 (Solution)
Hypothesis:

= There is no significant difference between methods A and B ().

= Method A is more effective than Method B ().

Level of significance: =0.01, one-tailed test


Test statistic: z statistic
Critical region: Reject if the computed Z>2.575.
Computation:

x1 - x2 78-70
z = = = 6.90
s12 s 22 4 6 2 2
Decision: We must reject + +may say that method A is
n1 ; nthat is, we

2 36 40
more effective than method B. The average grade of those with
method A is higher than the average grade of those who were
taught with method B.
Example 7
For a sample of 16 Philip light bulbs, the
mean lifetime is 1250 hours with a
standard deviation of 250 hours. For a
sample of Liwanag light bulbs, the mean
lifetime is 1400 hours with a standard
deviation of 300 hours. Use a 0.01 level
of significance to test whether or not
there is a significant difference between
the average lifetimes of the two brands
of light bulbs.
Example 7 (Solution)

Hypothesis:

= There is no significant difference between
the lifetimes of the two brands of light bulbs ().
= There is a significant difference between the
lifetimes of the two brands of light bulbs ().
Level of significance: =0.01, two-tailed test
Test statistic: t statistic with df= 16+12-2=26
Critical region: Reject if the computed t<-
2.779 or if t>2.779 .
Example 7 (Solution)
Computation:


(n1 - 1) s12 + (n2 - 1) s22 (16 - 1)(250) 2 + (12 - 1)(300) 2
sp = = = 272.277
n1 + n2 - 2 16 + 12 - 2

x1 - x2 1250-1400
t = = = 1.443
1 1 1 1
sp + 272.277 +
n1 n2 16 12
Decision: We must accept ; that is, we may say that the
average lifetimes of the two brands of light bulbs are the
same.
Example 8
To determine whether membership in a fraternity
is beneficial or detrimental to ones grades, the
following grade-point averages were collected
over a period of 5 years:
Year
1 2 3 4 5
Fraternity 2.0 2.0 2.3 2.1 2.4
Non-fraternity 2.2 1.9 2.5 2.3 2.4
Assuming the populations to be normal, test at
the 0.025 level of significance whether
membership in a fraternity is detrimental to
ones grades.
Example 8 (Solution)

Hypothesis:

or
or

Level of significance: =0.025, one-tailed test


Test statistic: t statistic
Critical region: Reject if the computed t<-
2.776, with df=5 -1=4
Example 7 (Solution)
Computation:


Fraterni Non- We
ty fraternity find that =0.5/5,
2.0 2.2 -0.2 0.04 and =0.14142.
2.0 1.9 0.1 0.01 Hence
d -0.1
2.3 2.5 -0.2 0.04 t = s = 0.14142 = -1.58
d
2.1 2.3 -0.2 0.04
n 5
2.4 2.4 0.0 0.00
-0.5 0.13

Decision: We must accept ; and conclude that membership


in a fraternity does not significantly affect ones grades.
Exercise 13
1. The average height of females in the
freshman class of a certain college has
historically been 162.5 centimeters with
a standard deviation of 6.9 centimeters.
Is there a 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 0.02 level of significance.
Exercise 13
According to a dietary study, high sodium intake may
be related to ulcers, stomach cancer, and migraine
headaches. The human requirement for salt is only
220 milligrams per day, which is surpassed in most
single servings of ready-to-eat cereals. If a random
sample of 20 similar servings of a certain cereal has a
mean sodium content of 244 milligrams and a
standard deviation of 24.5 milligrams, does this
suggest at the 0.05 level of significance that the
average sodium content for a single serving of such
cereal is greater than 220 milligrams? Assume the
distribution of sodium contents to be normal.
A study was conducted to see if increasing the substrate
concentration has an appreciable effect on the velocity of a
chemical reaction. With a substrate concentration of 1.5
moles per liter, the reaction was run 15 times, with an
average velocity of 7.5 micromoles per 30 minutes and a
standard deviation of 1.5. With a substrate concentration of
2.0 moles per liter, 12 runs were made, yielding an average
velocity of 8.8 micromoles per 30 minutes and a sample
standard deviation of 1.2. Is there any reason to believe
that this increase in substrate concentration causes an
increase in the mean velocity of the reaction of more than
0.5 micromole per 30 minutes? Use a 0.01 level of
significance and assume the populations to be
approximately normally distributed with equal variances.
A taxi company manager is trying to decide
whether the use of radial tires instead of
regular belted tires improves fuel economy.
Twelve cars were equipped with radial tires
and driven over a prescribed test course.
Without changing drivers, the same cars
were then equipped with regular belted tires
and driven once again over the test course.
The gasoline consumption, in kilometers per
liter, was recorded as follows:
Can we conclude that cars equipped with
radial tires give better fuel economy
than those equipped with belted tires?
Assume the populations to be normally
distributed.

Das könnte Ihnen auch gefallen