9 Testing the Hypothesis about the Population Parameter (1)

(9) TESTING OF HYPOTHESIS

A statistical hypothesis is a statement about a statistical population which on the basis of information obtained from
observed data, may or may not be true.
Objective
To determine if an assumption about some characteristic (parameter) of a population is supported by the information
obtained from the sample data.
Definitions
 The hypothesis being tested under the assumption that it is true, will be denoted by H 0 is called null hypothesis. An
alternative hypothesis denoted by H1 is the set of hypothesis which we accept when the null hypothesis H 0 is rejected.
 A hypothesis which completely specify the values of all the parameters of the distribution is called simple hypothesis. e.g.
In normal distribution with parameters  and 2, a hypothesis of the form  = 0 and 2 = 02 (where 0 and 02 are known
values) is a simple hypothesis.
 A hypothesis which does not specify the values of all the parameters of the distribution or a hypothesis which is not a
simple hypothesis is called composite hypothesis. e.g. if a variable is normally distributed with mean  = 0 and 2 > 02
(which is unspecified)
 The possible results of a sampling experiment can be divided into two groups. (i) results appearing consistent with the
hypothesis (ii) results leading us to reject the hypothesis. the first case is called acceptance region and second group is
called a rejection region (also called critical region)
 The probability of rejecting null hypothesis H0, when in fact it is true. The probability making this kind of error is equal to
the level of significance of the test. This is known as type I error. The probability of accepting the null hypothesis H 0 when in
fact it is false. This kind of error is known as type II error.

 Level of significance is the probability of rejecting the null hypothesis H0 when it is assumed to be true. i.e. the probability
of making a type I error. Level of significance is the risk of rejecting the null hypothesis H 0, when it is true. This probability is
equal to , the size of the test. The common choices of  are 0.05 and 0.01. If the cost of making a Type I error is high,
small values of α are preferred. If the cost of making a Type I error is not too high, larger values of α are typically used.
 A significance test is a statistical test lying down the procedure for deciding whether to accept or reject a statistical
hypothesis. A value is said to be significant when the probability of its occurrence under H 0 is equal to or less than
significance level . i.e. the value falls outside the acceptance region.
 A statistic on which the decision whether to accept or reject a null hypothesis is based is called a test statistic. e.g. the
student’s t-test, z-test or F distributions.
 A test for which the region of rejection is only located at one end, either on the left end or on the right end of the sampling
distribution of the test statistic is called a one tailed test or one sided test. When the region of rejection comprises both
ends or tails of the sampling distribution of the test statistic. The test is called two tailed test or two sided test.
 A two tailed test is appropriate when null hypothesis is H0 :  = 0 and the alternate hypothesis is H1 :  0. A left tailed
(lowered tailed) test is used if the hypotheses are H 0 :=0 and H1 : < 0.
 9 Testing the Hypothesis about the Population Parameter (2)

Testing Hypothesis about Mean of a Normal Distribution for Large Samples:

General Procedure:
(i) State the null hypothesis H0 with an alternate hypothesis H1
(ii) Level of significance  = 0
.
X 
(iii) State the test statistic to be used. i.e. Z = (for large sample n  30)
x
.
X 
t=
/ n
when  is unknown, use the S.D of the sampling distribution.
 N-n 
Also keep in mind, the use of x = N-1 or x = n

n
(iv) Select the region of rejection or critical region
(v) Compute the value of test statistic from the sample values
(vi) Decision or conclusion
Example 1
(a) Test the hypothesis that the mean weight of a population of people is 140 lbs. Using  = 15 lbs,  = 0.05 and a

sample of 36 people, find the value of X which would lead to reject the hypothesis.

(b) A random sample of n = 35 values gives X = 83. Can this sample be regarded as drawn from a normal population
with mean  = 80, and  = 7. (Sher M Ch)
Solution (a)
(i) we formulate our null hypothesis and alternate hypothesis as
H0 :  = 140 and H1 :   140
(ii) we see the level of significance is  = 0.05
.
X 
(iii) the test statistic to be used is Z =
/ n
(iv) for  = 0.05 is Z/2 = Z0.025 = ± 1.96 (called critical points)
Or region of rejection : |z|  1.96
i.e. z  -1.96 or z  1.96

(v) we calculate the value of X from the sample data as
.
X 
Z=
/ n

X140
 1.96 =
15/ 36
 
 X = 144.9 lbs and X = 135.1 lbs (called critical points)
 
(vi) Conclusion: the null hypothesis H0 :  = 140 lbs will be rejected if X > 144.9 lbs or X < 135.1 lbs
Solution (b)
(i) we formulate our null hypothesis and alternate hypothesis as
Null hypothesis : H0 :  = 80
and Alternative hypothesis: H1 :   80 (two sided)
(ii) we see the level of significance is  = 0.05
.
X 
(iii) the test statistic to be used is t =
/ n
9 Testing the Hypothesis about the Population Parameter (3)

(iv) critical points: t/2(n-1) = t0.025(24) =  2.064

or the critical region for  = 0.05 is t 2.064
( z  - 2.064 or z  2.064 ) [the meaning of |t|  2.064, is - 2.064  t  2.064
This called the region of acceptance]
(v) we calculate the value of z from the sample data as
.
X  83  80 35.9
t= = = 7 = 2.52
/ n 7 25
Conclusion:
Since our calculated value t = 2.52 falls in the critical region, so we reject our null hypothesis H 0 :  = 80 and
accept H1 :   80.

i.e. X = 83 cannot be regarded as drawn from the population with  = 80.
Example 2
Test the hypothesis that the mean of a normal population with known variance 70 is 31, if a

sample of size 13 gave X = 34. Let the alternate hypothesis be H1 :  > 31 and let  = 0.10
Solution
(i). we formulate our null hypothesis and alternate hypothesis as
H0 :  = 31 and H1 :  > 31 (one tailed test)
(ii). we are given the significance level as  = 0.10
.
X 
(iii). the test statistic to be used is t = ,which under null hypothesis has
/ n
a standard normal distribution.
(iv). The critical point is t/2(n-1) = t(0.05)(12) = 1.782
(v). We calculate the value of z from the sample data as
34  31 33.606
(vi). Calculation: t= = = 1.29
70/ 13 8.367
Conclusion.
Since the calculated value t = 1.29 falls in the region of acceptance, so we accept our null hypothesis H 0:  = 31.
Example 3
The marks obtained by students at a large number of colleges are known to be normally distributed with a mean of 25. a
random sample of 36 students showed an average number of marks of 27 with a standard deviation of 5. What conclusion should be
drawn?
Solution
(i) Since it is given that the average number of marks obtained by students at a large number of colleges is 25, we
therefore have our null hypothesis and alternate hypothesis as H0 :  = 25 and H1 :   25 (two tailed)
(ii) we specify the significance level as  = 0.05.
(iii) since  is not known and the sample size n = 36, we use standard deviation of the sampling distribution S = 5 in
.
X 
place of . Thus the test statistic is Z = , which has an approximate standard normal distribution under the
S/ n
given null hypothesis.
(iv) The region of rejection is z  1.96 (also called critical region)
 Either Z   1.96 or Z  1.96 (remember |z|  1.96 shows  1.96  z  1.96 )
Or critical points z/2 =  1.96
(v) We calculate the value of Z,
2725 26
Z= = = 2.40
5/ 36 5
9 Testing the Hypothesis about the Population Parameter (4)

(vi) Conclusion. Since the calculated value z = 2.40 falls in the region of rejection, we therefore reject our null
hypothesis and accept H1 :   25
Example 4
The breaking strength of cables produces by a manufacturer have mean 1800 lb and standard deviation 100 lb. By a new
technique in the manufacturing process it is claimed that the breaking strength can be increased. To test this claim a sample of 50
cables is tested and it is found that the mean breaking strength is 1850 lb. can we support the claim at a 0.01 level of significance?
Solution
H0:  = 1800 lb, and there is really no change in breaking strength.
H1:  > 1800 lb, and there is a change in breaking strength.
a one tailed test should be used here
Level of significance:  = 0.01
.
x 
Test of statistic to be used: z =
/ n
Critical point: Z = 2.33
1850  1800
Calculation: Z = = 3.55
100/ 50
Decision: As Z = 3.55 greater than 2.33, so the claim should be supported.
i.e. we reject H0.
Hence alternative is accepted.
i.e. . By a new technique in the manufacturing process, the breaking strength can be increased.
Example 5
A company manufactures rope whose breaking strength has a mean of 300 lbs and a standard deviation of 24 lbs. It is
believed that by a newly developed process the mean breaking strength can be increased.
(a) Design a decision rule for rejecting the old process at the 0.01 significance level if it is agreed to test 64 ropes.
(Schaum 10.13)
(b) Under the decision rule adopted in part (a), what is the probability of accepting the old process when in fact
the new process has increased the mean breaking strength to 310 lbs? Assume that the standard deviation is
still 24lbs.
Solution (a)
(i) Null Hypothesis: H0 :  = 300 lbs
Alternative Hypothesis: H1 :  > 300 lbs (right tailed test)
with n = 64 and  = 24 lbs
(ii) Level of significance:  = 0.01
(iii) Critical Points: z = 2.33
.
X 
(iv) Test statistic to be used: z =
/ n

(v) Calculation: X =  + Z / n = 300 + (2.33) (24)/ 64 = 307
i.e. observed sample mean
(vi) Conclusion: If the observed sample mean is greater than 307, we
reject H0 and we accept H1.
Solution (b) (The probability of making a type II error)
Null Hypothesis: H0 :  = 300 lbs
Alternative Hypothesis:
H1 :  = 310 lbs
with  = 24 lbs
9 Testing the Hypothesis about the Population Parameter (5)

The probability of accepting

H0 :  = 300 lbs when actually H1:  = 310 lbs is true
is the area under the right hand normal curve to the left of 307.
Now convert 307 in terms of z but with  = 310
.
X  307310
z= = = - 1.00
/ n 24/ 64
the corresponding area is 0.3413 (remember this area is from 310 to 307)
Hence area of shaded part is 0.5  0.3413 = 0.1587
i.e.  = P[type II error] = P[to accept null hypothesis when in fact it is false] = 0.1587
This is the probability of making a type II error.
Testing Hypothesis about Mean of a Normal Distribution for Small Samples:
Example 6
The price of a popular tennis racket at a national chain store is $179. A person bought five of the same racket at an online auction site
for the following prices:
155, 179, 175, 175, 161
Assuming that the auction prices of rackets are normally distributed, determine whether there is sufficient evidence in the
sample, at the 5% level of significance, to conclude that the average price of the racket is less than $179 if purchased at an online
auction.  = 5% = 0.05 , so /2 = 0.025, t/2(n-1) = t0.025(4) =  2.776
Solution
Step 1. The assertion for which evidence must be provided is that the average online price  is less than the average price in retail
stores, so the hypothesis test is
H0 : μ = 179
H1 : μ < 179 with level of significance  = 0.05
Step 2. The sample is small and the population standard deviation is unknown. Thus the test statistic is
.
X 
t=
s/ n
and has the Student t-distribution with n−1 = 5−1 = 4 degrees of freedom
Step 3. From the data we compute x̄ = 169 and s=10.39. Inserting these values into the formula for the test statistic gives
.
X  169  179
t= = =  2.152
s/ n 10.39/ 5
Step 4. Since the symbol in H1 is “<” this is a left-tailed test,
so there is a single critical value , −tα(n-1) = − t0.05(4) = −2.132.
The rejection region is (−∞, −2.132].
Step 5. As shown in Figure, the test statistic falls in the rejection region
. The decision is to reject H0.
In the context of the problem our conclusion is: The data provide sufficient evidence, at
the 5% level of significance, to conclude that the average price of such rackets purchased at online auctions is less than $179.
Example 7
Hinton press hypothesizes that the average life its largest web press is 14,500 hours. They know that the standard deviation
of press life is 2,100 hours. From a sample of 25 presses, the company finds a sample mean of 13,000 hours. At 0.01 significance level,
should the company conclude that the average life of the presses is less than the hypothesized 14,500 hours. (AIOU p-422)
Solution
(i) Null Hypothesis: H0 :  = 14500

Alternative Hypothesis: H1 :  < 14500 (lower tailed test) with  = 2100, n = 25 and X = 13000
(ii) Level of significance:  = 0.01
9 Testing the Hypothesis about the Population Parameter (6)

.
X 
(iii) Test Statistic to be used: t =
/ n
(iv) Critical Point: t()(n-1) =  2.33

(iv) Calculation: X =  + Z / n
2100
= 14500 + (-2.33)  = 13521.4 hours
25
Conclusion:

Observed sample mean X = 13000 is less than 13521.4, we should reject H0 and accept the alternative hypothesis i.e. the
average life of press is less than hypothesized value.
Example 8
American theatres know that a certain hit movie ran an average of 84 days in each city and the corresponding standard
deviation was 10 days. The manager of the Southeast district was interested in comparing the movie’s popularity in his region with that
in all of American’s other theatres. He randomly chose 75 theatres in his region and found that they ran the movie an average of 81.5
days. State appropriate hypothesis for testing whether there was a significant difference in the length of the movie’s run between
theatres in the Southeast district and all of American’s other theatres. At 1% level of significance, test these hypothesis.
Solution
Null Hypothesis: H0 :  = 84 days
Alternative Hypothesis: H1 :   84

with  = 10, n = 75 and X = 81.5
Level of significance:  = 0.01

Critical Points: X =  2.58
.
X 
Test Statistic to be used: z =
/ n

Calculation: X =  + Z/2 / n = 81.02 and 86.98

Conclusion: Because the observed sample mean X = 81.5 is in the acceptance region, so we accept H0.
Test of Hypothesis for  When n < 30
State the null hypothesis H0 with an alternate hypothesis H1
Level of significance
.
X 
State the test statistic to be used. i.e. t =
s/ n
where s the standard deviation of the sampling distribution.
Select the region of rejection or critical region
Compute the value of test statistic from the sample values
Decision or conclusion
Example 9
A test of the breaking strength of 6 ropes manufactured by a company showed a mean breaking strength of 7750 lb and a
standard deviation of 145 lb, whereas the manufacturer claim a mean breaking strength of 8000 lb. Can we support the manufacturer’s
claim at a level of significance of (a) 0.05 (b) 0.01?
Solution (a)
The null hypothesis H0 :  = 8000 lb, i.e. manufacturer’s claim
and alternate hypothesis H1 :  < 8000 lb (one tail test)
9 Testing the Hypothesis about the Population Parameter (7)

Level of significance :  = 0.05

.
X 
State the test statistic to be used. i.e. t =
s/ n
We have  = 8000 and s = 145
Select the critical points: t(n-1) = t(0.05)(5) =  2.01
.
X  7750  8000
Computation: t = = =  3.96
s/ n 145/ 5
Decision: as t =  3.96 is less than  2.01, so we reject Ho
Solution (a)
The null hypothesis H0 :  = 8000 lb, i.e. manufacturer’s claim
and alternate hypothesis H1 :  < 8000 lb (one tail test)
Level of significance :  = 0.01
.
X 
State the test statistic to be used. i.e. t =
s/ n
We have  = 8000 and s = 145
Select the critical points: t(n-1) = t(0.01)(5) =  3.36
.
X  7750  8000
Computation: t = = =  3.86
s/ n 145/ 5
Decision: as t =  3.86 is less than  3.36, so we reject Ho
Example 10
Given a sample mean of 83, a sample standard deviation of 12.5 and a sample size of 22, test the hypothesis that the value of
the population mean is 70 against the alternative that it is more than 70. Use the 0.025 significance level?
Solution

s = 12.5, n = 22, x = 83
Null Hypothesis:  = 70
Alternative Hypothesis:  > 70
Level of Significance:  = 0.025
Test Statistic to be used:
.
X 
t=
s/ n
Critical Points: As  = 0.025
t(0.025)(21) = 2.080
Calculation: Now convert t = 2.080 in terms of X
 s 12.5
x =  + t(0.025)(21) . = 70 + (2.080) = 75.54
n 22

Conclusion: because observed value x = 83 is greater than 75.54, so we reject H 0.
Exercise (1)
Ten individuals are chosen at random from a normal population and the height are found to be in inches 63, 63, 66, 67, 68, 69,
70, 70, 71 and 71. In the light of these data, discuss the suggestion that mean height in the population is 66 in. Set the level of
significance  = 0.05. (Sher II, p-250)
9 Testing the Hypothesis about the Population Parameter (8)

Exercise (2)
The masses in grams of 12 ball bearings taken at random from a lot are 31.4, 33.1, 35.9, 34.7, 33.4, 34.5, 35.0, 32.5, 36.9, 36.4,
35.8, 33.2. Test whether these values are consistent with a population mean of 34. Assume that the weights are normally distributed.
Example 11
Accept the hypothesis if the number of heads in a single sample of 100 tosses is between 40 and 60 inclusive. Reject the
hypothesis otherwise.
a) Find the probability of rejecting the hypothesis when it is actually correct. (Type I Error : the value of ).
b) What is the probability of accepting the hypothesis that the coin is fair when the actual probability of heads is p = 0.7.
(Type II Error: the value of )
Solution
We use normal approximation to binomial distribution if np > 5 and nq >5.
Now the mean and standard deviation are:
 = np = 100 ( ½ ) = 50 and  = npq = 100( ½ ) ( ½ ) = 5
Standardized values of 40 and 60 are – 2.0 and 2.0.
Area between 40 and 60 = 0.9544
H0 :  = 50, H1 :   50
Therefore probability of rejecting H0 is the probability
P[not getting heads between 40 and 60] = 1 – 0.9544 = 0.0456
Probability of accepting the hypothesis that the coin is fair when the actual probability of heads is p = 0.7 is highlighted in the
diagram.
To compute this area, we observe the distribution
under p = 0.7 which has
 = np = 70 and  = 100( 0.7 ) ( 0.3 ) = 4.58
corresponding standardized value of 60 is – 2.18
with corresponding area = 0.4854 (remember this area
is from 70 to 60 with  = 70 )
Hence Required Area =  = 0.5  0.4854 = 0.0146
Example 12
Consider the null hypothesis that the average weight of mail students in a certain college is 68 kg. A random sample of size 36
is drawn from a normal population with 3.6. Calculate the probabilities of committing type I and type II errors when testing the null
hypothesis against the alternative that   68 kilograms.
Type I Error
The probability of committing a type I error, or the level of significance of our test, is equal to the sum of the areas that have
been shaded in each tail of the distribution in Figure.

 
Therefore  = P(X < 67 when  = 68) + P(X > 69 when  = 68)
 
z-values corresponding to x = 67 and x = 69
9 Testing the Hypothesis about the Population Parameter (9)

67  68 69  68
z1 = =  1.67 and z2 = = 1.67
3.6 6 3.6 6
Therefore  = P(Z <  1.67) + P(Z > 1.67) = 2 P(Z <  1.67) = 0.950
Thus 95% of all samples of size 36 would lead us to reject  = 68 when in fact, it is true.
Type II Error
Now we evaluate  for various alternative hypotheses. If it is important to reject 0 when the true mean is some value  
70 or   66, then the probability of committing a type II error should be computed and examined for the alternatives  = 66 and  = 70.
Because of symmetry, it is only necessary to consider the probability of not rejecting the null hypothesis that  = 68 when the

alternative  = 70 is true. A type II error will result when the sample mean x falls between 67 and 69 when H1 is true. Therefore,
referring to Figure, we find that 


 = P( x  69 when  = 70).
 69  70
z-values corresponding to x = 69 when  = 70 is z = 0.45 =  2.22

Therefore  = P( Z < 2.22 ) = 0.5  0.4868 = 0.0132

Example 13
Air crew escape systems are powered by a solid propellant. The burning rate of this propellant is an important product
characteristic. Specifications require that the mean burning rate must be 50 centimeters per second. We know that the standard
deviation of burning rate is  = 2 centimeters per second. The experimenter decides to specify a type I error probability or significance

level of  0.05 and selects a random sample of n = 25 and obtains a sample average burning rate of x 513 centimeters per second.
What conclusions should be drawn?
(Ref. Book: “Applied Statistics and Probability for Engineers” 6th Edition, by Douglas Montgomary Example 9.2)
Solution
1. Parameter of interest: The parameter of interest is , the mean burning rate.
2. Null hypothesis: H0 : μ = 50 centimeters per second
3. Alternative hypothesis: H1 : μ ≠ 50 centimeters per second
4. Test statistic: The test statistic is
.
X 
z=
/ n
5. Critical Region or Critical points : The boundaries of the critical region would be z0.025 = 1.96 and –z0.025 = –1.96.

6. Computations: Because x 513 and  = 2,
51.3  50
z= = 3.25
2/ 25
7. Conclusion: Because the value z = 3.25 > 1.96 so we reject null hypothesis at the 0.05 level of significance.
Practical Interpretation: We conclude that the mean burning rate differs from 50 centimeters per second, based on a
sample of 25 measurements. In fact, there is strong evidence that the mean burning rate exceeds 50 centimeters per second.
9 Testing the Hypothesis about the Population Parameter (10)

Example 14
Consider above Example 13, Suppose that the true burning rate is 49 centimeters per second. What is β for the two-sided test
with α = 0.05, σ =2, and n = 25?
Solution
9 Testing the Hypothesis about the Population Parameter (11)

Table for Standard Normal Probabilities

9 Testing the Hypothesis about the Population Parameter (12)

Fisher and Yates Student’s t-Distribution

(v) d.f 0.25 0.10 0.05 0.025 0.01 0.005
1 1.000 3.078 6.314 12.706 31.821 63.657
2 0.816 1.886 2.920 4.303 6.965 9.925
3 0.765 1.838 2.353 3.182 3.541 5.841
4 0.741 1.533 2.132 2.776 3.747 4.604
5 0.727 1.476 2.015 2.571 3.365 4.032
6 0.718 1.440 1.943 2.447 3.143 3.707
7 0.711 1.415 1.895 2.365 2.998 3.499
8 0.706 1.397 1.860 2.306 2.896 3.355
9 0.703 1.383 1.833 2.262 2.821 3.250
10 0.700 1.372 1.812 2.228 2.764 3.169
11 0.697 1.363 1.796 2.201 2.718 3.106
12 0.695 1.356 1.782 2.179 2.681 3.055
13 0.694 1.350 1.771 2.160 2.650 3.012
14 0.692 1.345 1.761 2.145 2.624 2.977
15 0.691 1.341 1.753 2.131 2.602 2.947
16 0.690 1.337 1.746 2.120 2.583 2.921
17 0.689 1.333 1.740 2.110 2.567 2.898
18 0.688 1.330 1.734 2.101 2.552 2.878
19 0.688 1.328 1.729 2.093 2.539 2.861
20 0.687 1.325 1.725 2.086 2.528 2.845
21 0.686 1.323 1.721 2.080 2.518 2.831
22 0.686 1.321 1.717 2.074 2.508 2.819
23 0.685 1.319 1.714 2.069 2.500 2.807
24 0.685 1.318 1.711 2.064 2.492 2.797
25 0.684 1.316 1.708 2.060 2.485 2.787
26 0.684 1.315 1.706 2.056 2.479 2.779
27 0.684 1.314 1.703 2.052 2.473 2.771
28 0.683 1.313 1.701 2.048 2.467 2.763
29 0.683 1.311 1.699 2.045 2.462 2.756
30 0.683 1.310 1.697 2.042 2.457 2.750
40 0.681 1.303 1.684 2.021 2.423 2.704
60 0.679 1.296 1.671 2.000 2.390 2.660
120 0.677 1.289 1.658 1.980 2.358 2.617
 0.674 1.282 1.645 1.960 2.326 2.576