Attribution Non-Commercial (BY-NC)

5,6K vues

Attribution Non-Commercial (BY-NC)

- Hypothesis Testing
- Chapter 2 Hypothesis Testing
- QTM
- Null Hypothesis, Statistics, Z-Test, Significance
- Statistical Significance Hypothesis
- Steps in Hypothesis Testing
- QuantitativeTechniquesInManagement-MB007-Question.docx
- Hypothesis Testing Notes
- BusinessLaw MB012 Question
- 7. Confidence Intervals VIEW
- Z-test
- ADL 07 Quantitative Techniques in Management v4
- US-64
- ADL 07 Quantitative Techniques in Management AM4
- Assignement - QTM(Quantitive Techique Mgt)
- Student's_t-test
- Practice Exercises for Test of Hypothesis
- UTI-Dev
- Multiple Choice Questiones
- ADL-07-Ver1+

Vous êtes sur la page 1sur 37

the information contained in a random sample from that population.

population. It is essential that any sample is as representative as possible of the

population from which it is drawn.

Often, we may want to know things about populations but don’t have data for every

person or thing in the population. If a company’s customer service division wanted to

learn whether its customers were satisfied, it would not be practical or perhaps even

possible to contact every individual who purchased a product. Instead, the company

might select a sample or samples of the population.

For instance, we may be interested in the mean fill volume of a can of soft drink. The

mean fill volume in the population is required to be 300 millilitres. An Engineer takes a

random sample of 25 cans and computes the sample average to be x = 298 millilitres.

The Engineer will probably decide that the population mean is µ = 300 millilitres, even

though the sample was 298 millilitres because he or she knows that the sample mean is

a reasonable estimate of µ and that a sample mean of 298 millilitres is very likely to

occur, even if the true population mean is µ = 300 millilitres. In fact, if the true mean is

300 millilitres, tests of 25 cans made repeatedly, perhaps, every five minutes, would

produce values of x that vary both above and below µ = 300 millilitres.

The sample mean is a statistic; that is, a random variable that depends on the results

obtained in each particular sample. Since statistic is a random variable, it has a

probability distribution.

sample. Inferential statistics enables us to make an educated guess about a population

1

parameter based on a statistic computed from a sample randomly drawn from that

population.

formed when samples of size n are repeatedly taken from a population. If the sample

statistic is the sample mean, then the distribution is the sampling distribution of sample

means.

Suppose that a random sample of size n is taken from a normal population with mean µ

and variance σ 2 . Now each observation in this sample, say, X1, X2, ..., Xn, is a normally

and independently distributed random variable with mean µ and variance σ 2 . Then by

the reproductive property of normal distribution, the sample mean,

X1 + X 2 + + X n

X =

n

µ + µ ++ µ

µX = =µ

n

and variance

σ 2 + σ 2 + + σ 2 σ 2

σ X2 = =

n2 n

If we are sampling from a population that has an unknown probability distribution, the

sampling distribution of the sample mean will still be approximately normal with mean µ

σ2

and variance , if the sample size is large. This is one of the most useful theorems

n

called central limit theorem.

If X1, X2, ... , Xn is a random sample of size n taken from a population (finite or infinite)

with µ and finite variance σ 2 , and if X is the sample mean, the limiting form of the

distribution of

X −µ

Z = , as n → ∞, is the standard normal distribution.

σ/ n

2

NOMENCLATURE AND DEFINITIONS

Statistical Hypothesis

A statistical hypothesis is a statement about the parameters of one or more populations.

A null hypothesis is a hypothesis that might be falsified on the basis of observed data.

The null hypothesis typically proposes a general or default position, such as that there is

no relationship between two quantities, or that there is no difference between a

treatment and the control.

equality (≤, =, or ≤), it is the null hypothesis. If the original claim does not include

equality (<, not equal, >) then the null hypothesis is the complement of the original

claim. The null hypothesis always includes the equal sign.

Statement which is true if the null hypothesis is false. The type of test (left, right, or two-

tail) is based on the alternative hypothesis.

Test statistic

Sample statistic used to decide whether to reject or fail to reject the null hypothesis.

Critical region

Set of all values which would cause us to reject H0 (Region of rejection).

Critical value(s)

The value(s) which separate the critical region from the non-critical region. The critical

values are determined independently of the sample statistics.

The probability of rejecting the null hypothesis when it is true; α = 0.05 and α = 0.01 are

common. If no level of significance is given, use alpha = 0.05.

Decision

A statement based upon the null hypothesis. It is either "reject the null hypothesis" or

"fail to reject the null hypothesis". We will never accept the null hypothesis.

Conclusion

A statement which indicates at what level of significance the null hypothesis is rejected

or not rejected.

Statisticians will never accept the null hypothesis, we will fail to reject. In other words,

we'll say that it isn't, or that we don't have enough evidence to say that it isn't, but we'll

3

never say that it is, because someone else might come along with another sample

which shows that it isn't and we don't want to be wrong.

Type I error

Rejecting the null hypothesis when it is true. Usually the more serious error.

Type II error

Failing to reject the null hypothesis when it is false.

Alpha (α)

Probability of committing a Type I error.

Beta (β)

Probability of committing a Type II error.

Reject H0 Type I Error - α Correct Assessment

Fail to Reject H0 Correct Assessment Type II Error - β

Since Type I is the more serious error (usually), that is the one we concentrate on. We

usually pick alpha to be very small (0.05, 0.01). Notice here that, alpha is not a Type I

error. Alpha is the probability of committing a Type I error. Likewise beta is the

probability of committing a Type II error

TYPE OF TESTS

The type of test is determined by the Alternative Hypothesis (H1)

Notice the inequality points to the left

test statistic < critical value

4

Right Tailed Test

Notice the inequality points to the right

test statistic > critical value

Notice the inequality points to both sides

test statistic < critical value (left) or test statistic > critical value (right)

2. The second step is to consider the assumptions being made in doing the test;

for example, assumptions about the statistical independence or about the form

of the distributions of the observations. This is equally important as invalid

assumptions will mean that the results of the test are invalid.

3. Compute the relevant test statistic. The distribution of such a statistic under

the null hypothesis can be derived from the assumptions. In standard cases this

will be a well-known result. For example the test statistics may follow a

Student’s t distribution or a normal distribution. The distribution of the test

statistic partitions the possible values of the estimator into those for which the

null-hypothesis is rejected and those for which it is not.

4. Compare the test-statistic (S) to the relevant critical values (CV) (obtained from

tables in standard cases).

5. Decide to either fail to reject the null hypothesis or reject it in favor of the

alternative. The decision rule is to reject the null hypothesis (H0) if S > CV and

vice versa.

5

TESTS FOR MEAN

We will see how to conduct a test of hypothesis for a mean, when the following

conditions are met:

The sample is drawn from a normal or near-normal population.

the following conditions apply.

The sampling distribution is symmetric, unimodal, without outliers, and the

sample size is 15 or less.

The sampling distribution is moderately skewed, unimodal, without outliers, and

the sample size is between 16 and 40.

The sample size is greater than 40, without outliers.

This approach consists of four steps: (1) state the hypotheses, (2) formulate an analysis

plan, (3) analyze sample data, and (4) interpret results.

The table below shows three sets of hypotheses. Each makes a statement about how

the population mean, μ is related to a specified value μ0.

1 μ = μ0 μ ≠ μ0 2

2 μ > μ0 μ < μ0 1

3 μ < μ0 μ > μ0 1

The first case of hypotheses is an example of a two-tailed test, since an extreme value

on either side of the sampling distribution would cause to reject the null hypothesis. The

other two cases of hypotheses are one-tailed tests, since an extreme value on only one

side of the sampling distribution would cause a researcher to reject the null hypothesis.

6

TESTS FOR MEAN - KNOWN VARIANCE: (Z- Test)

Let X1, X2, ..., Xn is a random sample drawn from a normal population with known

variance σ2. Using sample data, conduct a one-sample Z-test.

σ/ n

where Z is the standard normal variable. Compare the Z calculated value with Z α and

Z α / 2 which are the critical values from normal distribution corresponding to α and α/2

probabilities representing one tail and two-tail tests. The decisions are given in the

following table.

Case

hypothesis hypothesis of tails Hypothesis

1 μ = μ0 μ ≠ μ0 2 Z ≥ Z α/ 2

2 μ > μ0 μ < μ0 1 Z ≤ −Z α

3 μ < μ0 μ > μ0 1 Z ≥ Zα

EXAMPLE:

Aircrew 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 σ = 2 centimeter per second. The experimenter decides to specify a

type I error probability or significance level of α = 0.05 and selects a random sample of

n= 25 and obtains a sample average burning rate of x = 51.3 centimeter per second.

What conclusion should be drawn?

We may solve this problem by following the eight-step procedure outlined as follows.

2. H0: μ = 50 centimeter per second

3. H1: μ ≠ 50 centimeter per second

4. α = 0.05

5. The test statistics is X − µ0

Z=

σ/ n

where we specified α = 0.05 and so the boundaries of the critical region are at

Z 0.025 =1.96 and −Z 0.025 = −1.96 from normal distribution tables.

7. Computations: Since x = 51.3 and σ = 2,

51 .3 − 50

Z0 = = 3.25

2 / 25

7

8. Conclusion: Since Z0 = 3.25 > 1.96 (Zα/2), we reject H0: μ = 50 at the 0.05

level of significance. Stated more completely, 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 is

not equal to 50 centimeters per second.

(i) when the alternative is of the type H1: μ > 50 centimeter per second the

conclusion would be to reject H0: μ ≤ 50 at the 0.05 level of significance,

since Z0 = 3.25 > 1.64 (Zα). Stated more completely, 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.

(ii) when the alternative is of the type H1: μ < 50 centimeter per second the

conclusion would be do not reject H0: μ ≥ 50 at the 0.05 level of

significance, since Z0 = 3.25 > - 1.64 (Zα). Stated more completely, 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.

Let X1, X2, … , Xn is a random sample drawn from a normal population with unknown

variance σ2. Using sample data, conduct a one-sample t-test.

s/ n

where t has a t-distribution with (n-1) degrees of freedom. Compare the t calculated

value with t α and tα / 2 which are the critical values from normal distribution

corresponding to α and α/2 probabilities representing one tail and two-tail tests. The

decisions are given in the following table.

Case

hypothesis hypothesis of tails Hypothesis

1 μ = μ0 μ ≠ μ0 2 t ≥tα / 2

2 μ > μ0 μ < μ0 1 t ≤ −tα

3 μ < μ0 μ > μ0 1 t ≥ tα

EXAMPLE

8

The increased availability of light materials with high strength has revolutionized the

design and manufacture of golf clubs, particularly drivers. Clubs with hollow heads and

very thin faces can result in much longer tee shots, especially for players of modest

skills. This is due partly to the “spring – like effect” that the thin face imparts to the ball.

Firing of golf ball at the head of the club and measuring the ratio of the outgoing velocity

of the ball to the incoming velocity can quantify this spring like effect. The ratio of

velocities is called the coefficient of restitution of the club. An experiment was

performed in which 15 drivers produced by a particular club maker were selected at

random and their coefficients of restitution measured. In the experiment the golf balls

were fired from air cannon so that the incoming velocity and spin rate of the ball could

be precisely controlled. It is of interest to determine if there is evidence (with α = 0.05) to

support a claim that the mean coefficient of restitution exceeds 0.82. The observations

are:

0.8411 0.8191 0.8182 0.8125 0.8750

0.8580 0.8532 0.8483 0.8276 0.7983

0.8042 0.8730 0.8282 0.8359 0.8660

We may solve this problem by following the eight-step procedure outlined as follows.

2. H0: μ = 0.82

3. H1: μ > 0.82. We want to reject H0 if the mean coefficient of restitution

exceeds 0.82

4. α = 0.05

5. The test statistics is X − µ0

t0 =

s/ n

7. Computations: Since x = 0.83725, s = 0.02456, μ0 = 0.82 and n = 15 we

have,

0.83725 − 0.82

t0 = = 2.72

0.02456 / 15

8. Conclusion: Since t0 = 2.72 > 1.761 (t0.05, 14), we reject H0: μ = 0.82 at the

0.05 level of significance that the mean coefficient of restitution exceeds 0.82

based on a sample of 15 measurements.

i. when the alternative is of the type H1: μ ≠ 0.82 the conclusion would be to

reject H0: μ = 0.82 at the 0.05 level of significance, since t0 = 2.72 > 2.145 (t0.025). Stated

more completely, we conclude that the mean coefficient of restitution is not equal to

0.82 based on a sample of 15 measurements

9

ii. when the alternative is of the type H1: μ < 0.82, the conclusion would be

do not reject H0: μ = 0.82 at the 0.05 level of significance, since t0 = 2.72 > - 1.761 (

-t0.05, 14),. Stated more completely, we conclude that the mean coefficient of restitution

exceeds 0.82 based on a sample of 15 measurements

PROBLEMS:

claims that the engine will run continuously for 5 hours (300 minutes) on a single

gallon of regular gasoline. Suppose a simple random sample of 50 engines is

tested. The engines run for an average of 295 minutes, with a standard deviation

of 20 minutes. Test the null hypothesis that the mean run time is 300 minutes

against the alternative hypothesis that the mean run time is not 300 minutes. Use

a 0.05 level of significance. (Assume that run times for the population of engines

are normally distributed.)

2. Bon Air Elementary School has 300 students. The principal of the school thinks

that the average IQ of students at Bon Air is at least 110. To prove her point, she

administers an IQ test to 20 randomly selected students. Among the sampled

students, the average IQ is 108 with a standard deviation of 10. Based on these

results, should the principal accept or reject her original hypothesis? Assume a

significance level of 0.01.

10

TESTS FOR A POPULATION PROPORTION

Suppose that a random sample of size ‘n’ is drawn from a large population and that

X(≤n) observations in this sample belong to a specified class of interest. Then P̂ = X/n

is a point estimator of the proportion ‘p’ of the population that belongs to this class. Note

that n and p are parameters of binomial distribution. When n is relatively large and if p

is not too close to either 0 (zero) or 1 (one), then P̂ is approximately normal with mean

p and variance p(1- p)/n. For this approximation we require that np and n(1-p) be

greater than 5.

1 p= p0 p≠ p0 2

2 p > p0 p < p0 1

3 p < p0 p > p0 1

Let X be the number of observations in a random sample of size ‘n’ that belongs to the

class associated with p. Using sample data, conduct a one-sample Z-test as follows:

Z=

p 0 (1 − p 0 ) / n

where Z has the standard normal distribution. Compare the Z calculated value with Z α

and Z α / 2 which are the critical values from normal distribution corresponding to α and

α/2 probabilities representing one tail and two-tail tests. The decisions are given in the

following table.

Case

hypothesis hypothesis of tails Hypothesis

1 p = p0 p ≠ p0 2 Z ≥ Z α/ 2

2 p > p0 p < p0 1 Z ≤ −Z α

3 p < p0 p > p0 1 Z ≥ Zα

The first case of hypotheses is an example of a two-tailed test, since an extreme value

on either side of the sampling distribution would cause to reject the null hypothesis. The

other two cases of hypotheses are one-tailed tests, since an extreme value on only one

side of the sampling distribution would cause a researcher to reject the null hypothesis.

EXAMPLE:

applications. The customer requires that the process fraction defective at a critical

11

manufacturing step not exceed 0.05 and that the manufacturer demonstrate process

capability at this level of quality using α = 0.05. The semiconductor manufacturer takes

a random sample of 200 devices and finds that 4 of them are defective. Can the

manufacturer demonstrate process capability for the customer?

We may solve this problem by following the eight-step procedure outlined as follows.

2. H0: p = 0.05

3. H1: p < 0.05; (this formulation of the problem will allow the

manufacturer to make a strong claim about process capability if null hypothesis

H0: p = 0.05 is rejected).

4. α = 0.05

5. The test statistics is Pˆ − p 0

Z0 =

p 0 (1 − p 0 ) / n

specified α = 0.05 and so the boundary of the critical region is −Z 0.025 = −1.96

from normal distribution tables.

7. Computations: Since X=4, n=200, p0= 0.05 and P̂ = 0.02

0.02 − 0.05

Z0 = = −1.95

0.05 (1 − 0.95 ) / 200

the 0.05 level of significance and conclude that the process fraction defective is

less than 0.05. Hence the process is capable.

PROBLEMS:

1. The CEO of a large electric utility claims that 80 percent of his 1,000,000

customers are very satisfied with the service they receive. To test this claim,

the local newspaper surveyed 100 customers, using simple random sampling.

Among the sampled customers, 73 percent say they are very satisfied. Based

on these findings, can we reject the CEO's hypothesis that 80% of the

customers are very satisfied? Use a 0.05 level of significance.

2. The previous problem is stated a little bit differently. Suppose the CEO claims

that at least 80 percent of the company's 1,000,000 customers are very

satisfied. Again, 100 customers are surveyed using simple random sampling.

The result: 73 percent are very satisfied. Based on these results, should we

accept or reject the CEO's hypothesis? Assume a significance level of 0.05.

12

TESTS FOR DIFFERENCE BETWEEN TWO POPULATION PROPORTIONS

We now consider the case where there are two binomial parameters of interest, say p1

and p2. Suppose that two independent random samples of sizes n1 and n2 are taken

from two populations, and let X1 and X2 represent the number of observations that

belong to the class of interest in samples or 1 and 2, respectively. Let,

X , X X + X2

Pˆ1 = 1 Pˆ2 = 2 and Pˆ = 1

n1 n2 n1 + n 2

p1 = p2 p1 ≠ p2

1 or or 2

p1 - p2 = 0 p1 - p2 ≠ 0

p1 ≥ p2 p1 < p2

2 or or 1

p1 - p2 ≥ 0 p1 - p2 < 0

p1 ≤ p2 p1 > p2

3 or or 1

p1 - p2 ≤ 0 p1 - p2 > 0

Z=

p1 (1 − p1 ) p 2 (1 − p 2 )

+

n1 n2

Pˆ1 − Pˆ2

Z0 =

1 1

Pˆ (1 − Pˆ ) +

n1 n2

Compare the calculated value of Z0 with Z α and Z α / 2 which are the critical values from

normal distribution corresponding to α and α/2 probabilities, representing one tail and

two-tail tests. The decisions are given in the following table.

Case

hypothesis hypothesis of tails Hypothesis

1 p1 = p2 p1 ≠ p2 2

13

or or Z ≥ Z α/ 2

p1 - p2 = 0 p1 - p2 ≠ 0

p1 ≥ p2 p1 < p2

2 or or 1 Z ≤ −Z α

p1 - p2 ≥ 0 p1 - p2 < 0

p1 ≤ p2 p1 > p2

3 or or 1 Z ≥ Zα

p1 - p2 ≤ 0 p1 - p2 > 0

EXAMPLE

Suppose the Acme Drug Company develops a new drug, designed to prevent colds.

The company states that the drug is equally effective for men and women. To test this

claim, they choose a simple random sample of 100 women and 200 men from a

population of 100,000 volunteers. At the end of the study, 38% of the women caught a

cold; and 51% of the men caught a cold. Based on these findings, can we reject the

company's claim that the drug is equally effective for men and women? Use a 0.05 level

of significance.

We may solve this problem by following the eight-step procedure outlined as follows.

This is a two tail test.

1. The parameters of interest are p1 and p2 the proportion of men and women who

caught a cold.

. H0: p1 = p2

. H1: p1 ≠ p2

. α = 0.05

. The test statistics is

Pˆ1 − Pˆ2

Z0 =

1 1

Pˆ (1 − Pˆ ) +

n1 n 2

where, Pˆ1 = 0.51 , Pˆ2 = 0.38 and Pˆ = 0.5933 , n1=200 and n2 = 100

6. Reject H0 if Z ≥ 1.96 or if Z ≤ -1.96. Note that this results from step 4, where we

specified α = 0.05 and so the boundaries of the critical region are at

Z 0.025 =1.96 and −Z 0.025 = −1.96 from normal distribution tables.

14

0.51 − 0.38

Z0 = = 2.17

1 1

0.5933 (1 − 0.5933 ) +

200 100

8. Conclusion: Since Z0 = 2.17 > 1.96 (Zα/2), we reject H0: p1 = p2 at the 0.05 level

of significance. Hence we reject the claim that the drug is equally effective for

men and women.

i. when the alternative is of the type H 1: p1 > p2 the conclusion would be to reject

H0: p1 = p2 at the 0.05 level of significance, since Z0 = 2.17 > 1.64 (Zα) and

conclude that the drug is more effective for men than women.

ii. when the alternative is of the type H1: p1 < p2 the conclusion would be do not

reject H0: p1 = p2 at the 0.05 level of significance, since Z0 = 2.17 > - 1.64 (Zα)

and conclude that the drug is more effective for men than women.

Let us see how to conduct a hypothesis test for the difference between two means from

two independent populations with means μ1 and μ2 and variances σ12 and σ22

respectively. Inferences will be based on two random samples of sizes n1 and n2

respectively.

Let X11, X12, … , X1n1 is a random sample of n1 observations from the population with

mean μ1 and variance σ12 and X21, X22, … , X2n2 is a random sample of n2 observations

from the population with mean μ2 and variance σ22. Assume that both populations are

independent and normal.

X 1 − X 2 − ( µ1 − µ 2 )

Z=

σ 12 σ 22

+

n1 n2

We now consider hypothesis testing on the difference in the means, μ1 - μ2. The various

hypotheses are stated in the following table.

15

μ1 = μ 2 μ1≠ μ2

1 or or 2

μ1 - μ2 =0 μ1 - μ2 ≠0

μ1 ≥ μ2 μ1 < μ2

2 or or 1

μ1 - μ2 ≥0 μ1 - μ2 <0

μ1 ≤ μ2 μ1 > μ2

3 or or 1

μ1 - μ2 ≤ 0 μ1 - μ2 > 0

Given the two sample observations, calculate the sample means x1 and x 2 . Under

the null hypothesis the test statistic becomes,

x1 − x 2

Z0 =

σ 12 σ 22

+

n1 n 2

Compare the calculated value of Z0 with Z α and Z α / 2 which are the critical values from

normal distribution corresponding to α and α/2 probabilities, representing one tail and

two-tail tests. The decisions are given in the following table.

hypothesis hypothesis tails Hypothesis

μ1 = μ 2 μ1≠ μ2

1 or or 2 Z 0 ≥ Zα/ 2

μ1 - μ2 =0 μ1 - μ2 ≠0

μ1 ≥ μ2 μ1 < μ2

2 or or 1 Z 0 ≤ −Z α

μ1 - μ2 ≥0 μ1 - μ2 <0

μ1 ≤ μ2 μ1 > μ2

3 or or 1 Z 0 ≥ Zα

μ1 - μ2 ≤ 0 μ1 - μ2 > 0

EXAMPLE:

A product developer is interested in reducing the drying time of primer paint. Two

formulations of the paint are tested. Formulation-1 is the standard chemistry and

Formulation-2 has a new drying ingredient that should reduce the drying time. From

experience it is known that the standard deviation of drying time is 8 minutes and this

inherent variability should be unaffected by the addition of the new ingredient. Ten

specimens are painted with Formulation-1 and another 10 specimens are painted with

16

Formulation-2; the 20 specimens are painted in random order. The two sample average

drying times are x1 = 121 minutes and x 2 = 112 minutes respectively. What

conclusions can be the product developer draw about the effectiveness of the new

ingredient, using α = 0.05.

We may solve this problem by following the eight-step procedure outlined as follows.

μ2

2. H0: μ 1 - μ 2 = 0, (or) μ 1 = μ 2

3. H1: μ 1 - μ 2 > 0, (or) μ 1 > μ 2. We want to reject H0 if the new

ingredient reduces the drying time.

4. α = 0.05

5. The test statistics is

x1 − x 2

Z0 =

σ 12 σ 22

+

n1 n2

6. Reject H0 if Z0 ≥ 1.645 (Zα). Note that this results from step 4, where

we specified α = 0.05 and so the boundaries of the critical region are at

Z 0.05 =1.645 from normal distribution tables.

7. Computations: Since x1 =121 minutes, x 2 =112 minutes, σ12 =σ22 =

82 = 64 minutes and n1= n2 = 10, the value of the test statistics is,

121 − 112

Z0 = = 2.52

82 82

+

10 10

0.05 level of significance and conclude that adding the new ingredient to the

paint significantly reduces the drying time.

Note:

(i) In case of a two tail test – when the alternative hypothesis is of the type H1:

μ 1 ≠ μ2, (or) H1: μ 1 - μ 2 ≠ 0 the conclusion would be to reject H0: μ 1 - μ 2 = 0

at the 0.05 level of significance, since Z0 = 2.52 > 1.96 (Zα/2). That is the

mean drying time is significantly different for the two types of primers.

(ii) In case of an one tail test – when the alternative hypothesis is of the

type H1: μ 1 < μ 2 (or), H1: μ 1 - μ 2 < 0, the conclusion would be do not reject

H0: μ 1 - μ 2 = 0 at the 0.05 level of significance, since Z0 = 2.52 > - 1.645

and conclude that adding the new ingredient to the paint significantly reduces

the drying time.

17

Let us see how to conduct a hypothesis test for the difference between two means from

two independent populations with means μ1 and μ2 and unknown variances σ12 and

σ22 respectively. Here it is assumed that σ12 = σ22 = σ2. That is the variances of the

two normal populations are unknown but are equal. Inferences will be based on two

random samples of sizes n1 and n2 respectively.

Let X11, X12, … , X1n1 is a random sample of n1 observations from the population with

mean μ1 and X21, X22, … , X2n2 is a random sample of n2 observations from the

population with mean μ2 and common unknown variance σ2. Assume that both

populations are independent and normal.

(n1 − 1) s12 + (n 2 − 1) s 22 1 1

s 2p = , where s1 = n −1 ∑( x1i − x1 ) and s 2 = n − 1 ∑( x 2i − x 2 )

2 2 2 2

n1 + n 2 − 2 1 2

X 1 − X 2 − ( µ1 − µ 2 )

t=

1 1

sp +

n1 n 2

We now consider hypothesis testing on the difference in the means, μ1 - μ2. The various

hypotheses are stated in the following table.

μ1 = μ 2 μ1≠ μ2

1 or or 2

μ1 - μ2 =0 μ1 - μ2 ≠0

μ1 ≥ μ2 μ1 < μ2

2 or or 1

μ1 - μ2 ≥0 μ1 - μ2 <0

18

μ1 ≤ μ2 μ1 > μ2

3 or or 1

μ1 - μ2 ≤ 0 μ1 - μ2 > 0

Given the two sample observations, calculate the sample means x1 , x 2 and the

pooled estimate sp. Under the null hypothesis the test statistic becomes,

x1 − x 2

to =

1 1

sp +

n1 n2

Compare the calculated value of t0 with t α and tα / 2 which are the critical values from

normal distribution corresponding to α and α/2 probabilities, representing one tail and

two-tail tests. The decisions are given in the following table.

hypothesis hypothesis tails Hypothesis

μ1 = μ 2 μ1≠ μ2 t 0 ≥ tα / 2

1 or or 2

μ1 - μ2 =0 μ1 - μ2 ≠0

μ1 ≥ μ2 μ1 < μ2 t 0 ≤ −tα

2 or or 1

μ1 - μ2 ≥0 μ1 - μ2 <0

μ1 ≤ μ2 μ1 > μ2 t 0 ≥ tα

3 or or 1

μ1 - μ2 ≤ 0 μ1 - μ2 > 0

EXAMPLE:

Two catalysts are being analyzed to determine how they affect the mean yield of a

chemical process. Specifically, catalyst-1 is currently in use, but catalyst-2 is

acceptable. Since catalyst-2 cheaper, it should be adopted, if it does not change the

process yield. A test is run in the pilot plant and the results are shown in the table

below. Is there any difference between the mean yields? Use α = 0.05, and assume

equal variances.

1 91.50 89.19

2 94.18 90.95

3 92.18 90.46

4 95.39 93.21

5 91.79 97.19

6 89.07 97.04

7 94.72 91.07

8 89.21 92.75

19

From the data it can be calculated that x1 = 92.255, x 2 = 92.733, s1 = 2.39 and s2 = 2.98.

The problem is solved using the eight-step hypothesis testing procedure as follows.

μ 2=0.

2. H0: μ 1 - μ 2 = 0, (or) μ 1 = μ 2

3. H1: μ 1 - μ 2 ≠ 0, (or) μ 1 ≠ μ 2.

4. α = 0.05

5. The test statistics is x1 − x 2

to =

1 1

sp +

n1 n2

6. Reject H0 if t0 > 2.145 ( = t0.025,14) (or) if t0 < -2.145 (= -t0.025,14). Note that

this results from step 4, where we specified α = 0.05 and so the boundaries of

the critical region are at t0.025,14 = 2.145 from t-distribution tables (two tail).

7. Computations: We have x1 = 92.255, x 2 = 92.733, s1 = 2.39, s2 =

2.98 and n1= n2 = 8. Therefore,

(n − 1) s12 + (n 2 − 1) s 22 7(2.39) 2 + 7(2.98) 2

s 2p = 1 = = 7.30

n1 + n2 − 2 8+8−2

92 .255 − 92 .733

t0 = = −0.35

1 1

2.70 +

8 8

we do not reject H0: μ 1 - μ 2 = 0 at the 0.05 level of significance and conclude

that there is no strong evidence that catalyst-2 results in a mean yield that differs

from the mean yield when catalyst-1 is used.

20

i. when the alternative is of the type H1: μ1 - μ2 > 0 (or) H1: μ1 > μ2 the

conclusion would be do not reject H0: μ 1 - μ 2 = 0, (or) μ 1 = μ 2 at the 0.05 level of

significance, since t0 = -0.35 < 1.761 (=t0.05,14). We conclude that the mean yield of

Catalyst-1 is not significantly greater than the mean yield when catalyst-2 is used.

ii. when the alternative is of the type H1: μ1- μ2 < 0 (or) H1: μ1 < μ2 the

conclusion would be do not reject H0: μ 1 - μ 2 = 0, (or) μ 1 = μ 2 at the 0.05 level of

significance, since t0 = -0.35 > -1.761 (= -t0.05,14). We conclude that the mean yield of

Catalyst-1 is not significantly less than the mean yield when catalyst-2 is used.

PROBLEMS

1. Within a school district, students were randomly assigned to one of two Math

teachers - Mrs. Smitha and Mrs. Lakshmi. After the assignment, Mrs. Smitha had 30

students, and Mrs. Lakshmi had 25 students. At the end of the year, each class took

the same standardized test. Mrs. Smitha’s students had an average test score of 78,

with a standard deviation of 10; and Mrs. Lakshmi 's students had an average test

score of 85, with a standard deviation of 15. Test the hypothesis that Mrs. Smitha

and Mrs. Lakshmi are equally effective teachers. Use a 0.10 level of significance.

(Assume that student performance is approximately normal.)

2. The Acme Company has developed a new battery. The engineer in charge claims

that the new battery will operate continuously for at least 7 minutes longer than the

old battery. To test the claim, the company selects a simple random sample of 100

new batteries and 100 old batteries. The old batteries run continuously for 190

minutes with a standard deviation of 20 minutes; the new batteries, 200 minutes with

a standard deviation of 40 minutes. Test the engineer's claim that the new batteries

run at least 7 minutes longer than the old. Use a 0.05 level of significance.

21

PAIRED t-TEST

The paired t-test is generally used when measurements are taken from the same

subject before and after some manipulation such as injection of a drug. For example, ya

paired t test can be used to determine the significance of a difference in blood pressure

before and after administration of an experimental drug. Paired t-test may also be used

to compare samples that are subjected to different conditions, provided the samples in

each pair are identical otherwise. For example, we might test the effectiveness of a

water additive in reducing bacterial numbers by sampling water from different sources

and comparing bacterial counts in the treated versus untreated water sample. Each

different water source would give a different pair of data points.

The number of points in each data set must be the same, and they must be organized in

pairs, in which there is a definite relationship between each pair of data points. Clearly

for paired t-test, the data is dependent, i.e. there is a one-to-one correspondence

between the values in the two samples. For example, same subject measured before

and after a process change or same subject measured at different times.

Let (X11, X21), (X12, X22), … , (X1n, X2n) be a set of n paired observations of a sample

drawn from two populations with means μ1 and μ2 and variances σ12 and σ22 respectively.

Define the differences between each pair of observations as Dj = X1j - X2j, j = 1,2, … , n.

Then Dj’’s are assumed to be normally distributed with mean μD = μ1 - μ2 and variance

σD2. Hence testing hypothesis about the difference between μ1 and μ2 can be

accomplished by performing a one-sample t-test on μD.

D − µD

Then, t D = has a t-distribution with (n-1) degrees of freedom. An estimator of

σD / n

1 1

σD2 is given by s D =

2

n −1

∑(di − d ) 2 where di = x1j-x2j and d = ∑d i

n

We now consider hypothesis testing on the difference in the means, μ1 - μ2. The various

hypotheses are stated in the following table.

Case Null

Alternative hypothesis Number of tails

hypothesis

1 μD = 0 μD ≠ 0 2

2 μD ≥ 0 μD < 0 1

3 μD ≤ 0 μD > 0 1

Given the pairs of sample observations, calculate d and sD2. Under the null hypothesis

the test statistic becomes,

d

t0 =

sD / n

22

where t0 has a t-distribution with (n-1) degrees of freedom. Compare the t0 calculated

value with tα and tα / 2 which are the critical values from normal distribution corresponding

to α and α/2 probabilities representing one tail and two-tail tests. The decisions are

given in the following table.

Case

hypothesis hypothesis of tails Hypothesis

1 μD = 0 μD ≠ 0 2 t 0 ≥ tα / 2

2 μD ≥ 0 μD < 0 1 t 0 ≤ −tα

3 μD ≤ 0 μD > 0 1 t 0 ≥ tα

EXAMPLE:

The following data refers to Strength predictions for nine Steel Plate Girders by

Karlsruhe and Lehigh Methods. Test whether there is any significant difference between

the two methods.

1 1.186 1.061 0.119

2 1.151 0.992 0.159

3 1.322 1.063 0.259

4 1.339 1.062 0.277

5 1.200 1.065 0.138

6 1.402 1.178 0.224

7 1.365 1.037 0.328

8 1.537 1.086 0.451

9 1.559 1.052 0.507

The problem is solved using the eight-step hypothesis testing procedure as follows.

between the two methods, say, μD = μ1 - μ 2 = 0.

2. H0: μD = 0

3. H1: μD ≠ 0

4. α = 0.05

5. The test statistics is

d

t0 =

sD / n

6. Reject H0 if t0 > 2.306 ( = t0.025,8) (or) if t0 < -2.306 (= -t0.025,8). Note that

this results from step 4, where we specified α = 0.05 and so the boundaries of

the critical region are at t0.025,8= 2.306 from t-distribution tables (two tail).

23

7. Computations: We have d = 0.2736, and sD = 0.1356 and n = 9.

Therefore,

0.2736

t0 = = 6.05

0.1356 9

the 0.05 level of significance and conclude that the strength prediction methods

yield different results.

iii. when the alternative is of the type H1: μD > 0 the conclusion would be to

reject H0: μD = 0, at the 0.05 level of significance, since t0 = 6.05 > 1.860 (=t0.05,8).

Specifically, the data indicate that the Karlsruhe Method produces, on the average

higher strength predictions than does the Lehigh Methods.

iv. when the alternative is of the type H1: μD < 0 the conclusion would be do

not reject H0: μD = 0, at the 0.05 level of significance, since t 0 = 6.05 >

-1.860 (=-t0.05,8). Specifically, the data indicate that the Karlsruhe Method produces, on

the average higher strength predictions than does the Lehigh Methods.

Suppose that we wish to test the hypothesis that the variance of a normal population σ 2

equals a specified value, say σ02 or equivalently, that the standard deviation σ is equal

σ0. Let X1, X2, … , Xn be a random sample of n observations from this population.

The table below shows three sets of hypotheses for testing the variance. Each makes a

statement about how the population variance, σ2 is related to a specified value σ02.

1 σ2 = σ02 σ2 ≠σ02 2

2 σ2 ≥ σ02 σ2 < σ02 1

3 σ2 ≤ σ02 σ2 > σ02 1

( n − 1) S 2

We use the test statistic χ =

2

.

σ 02

( n − 1) S 2

Under the null hypothesis H0: σ = σ0 , the statistic χ =

2 2 2

has chi-square

σ 02

distribution with (n-1) degrees of freedom.

24

Given a sample of observations calculate, the sample variance

1 .

s2 = ∑

(x − x)2

(n −1)

i

(n − 1) s 2

χ 02 =

σ 02

where χ02 has a chi-square distribution with (n-1) degrees of freedom. Compare the χ02

calculated value with χα2 and χ 2 which are the critical values from chi-square

α/2

distribution corresponding to α and α/2 probabilities representing one tail and two-tail

tests. The decisions are given in the following table.

Case

hypothesis hypothesis of tails Hypothesis

χ02 ≥ χ α / 2

2

1 σ2 = σ02 σ2 ≠σ02 2 or

χ ≤ χ 1−α / 2

2 2

0

χ02 ≤ χ1−α

2

2 σ2 ≥ σ02 σ2 < σ02 1

3 σ2 ≤ σ02 σ2 > σ02 1 χ02 ≥ χα2

EXAMPLE:

An automatic filling machine is used to fill bottles with liquid detergent. A random

sample of 20 bottles results in a sample variance of fill volume of s2 =0.0153 (fluid

ounces)2. If the variance of fill volume exceeds 0.01 (fluid ounces)2, an unacceptable

portion of bottles will be underfilled or overfilled. Is there evidence in the sample data to

suggest that the manufacturer has a problem with underfilled or overfilled bottles?

Use α = 0.05 and assume that fill volume has a normal distribution.

The problem is solved using the eight-step hypothesis testing procedure as follows.

2. H0: σ2 = 0.01

3. H1: σ2 > 0.01

4. α = 0.05

5. The test statistics is 2 (n − 1) s 2

χ0 =

σ 02

6. Reject H0 if χ02 > 30.14 (= χ02.05 ,19 ). Note that this results from step 4, where we

specified α = 0.05 and so the critical region is at χ02.05 ,19 = 30.14 from chi-

square distribution tables (one tail).

25

7. Computations: We have s2 =0.0153. Therefore,

(19 (0.0153 )

χ02 = = 29 .07

0.01

8. Conclusion: Since χ02 = 29.07 < 30.14 ( = χ02.05 ,19 ) we do not reject the null

hypothesis H0: σ2 = 0.01 at the 0.05 level of significance and conclude that

there is no strong evidence that the variance of fill volume exceeds 0.01(fluid

ounces)2.

Note: (i) In case of a two tail test – when the alternative hypothesis is of the

type H1: σ2 ≠ 0.01 the conclusion would be do not reject H0: σ2 = 0.01

at the 0.05 level of significance since, χ 2 = 29.07 < 32.85 ( χ 2 ) and

0 α/2

2 2

0

the type H1: σ2 < 0.01, the conclusion would be do not reject the null

hypothesis H0: σ2 = 0.01 at the 0.05 level of significance, since

χ02 = 29.07 > 10.12 (= χ02.05 ,19 ) and conclude that there is no strong

evidence that the variance of fill volume exceeds 0.01(fluid ounces)2.

Suppose that two independent normal populations are of interest, where the population

means and variances, say, μ1, σ12 , μ2, and σ22 are unknown. We wish to test the

hypothesis about the equality of two variances, say, H 0: σ12 = σ22. Assume that two

random samples of sizes n1 and n2 from the two populations respectively, and let s12 and

s22 be the respective samples variances based on the two samples.

26

The null and alternative hypotheses are given in the following table.

1 σ12 = σ22 σ12 ≠ σ22 2

2 σ12 ≥ σ22 σ12 < σ22 1

3 σ12 ≤ σ22 σ12 > σ22 1

Let X11, X12, … , X1n1 is a random sample of n1 observations from the population with

mean μ1 and variance σ12 and X21, X22, … , X2n2 is a random sample of n2 observations

from the population with mean μ2 and variance σ22. Assume that both populations are

independent and normal. Let s12 and s22 be the respective samples variances based on

the two samples.

σ 22 s12

Define the ratio, F = . This F statistic has F-distribution with (n1-1) numerator

σ 12 s 22

degrees of freedom and (n2-1) denominator degrees of freedom.

s12

Under the hull hypothesis the test statistic becomes, F0 = 2 .

s2

Compare the calculated value of F0 with Fα and Fα/2 which are the critical values from F-

distribution against [(n1-1), (n2-1)] degrees of freedom corresponding to α and α/2

probabilities representing one tail and two-tail tests. The decisions are given in the

following table.

Case

hypothesis hypothesis of tails Hypothesis

F0 ≥ Fα/2

1 σ12 = σ22 σ12 ≠ σ22 2 or

F0 ≤ F1-α/2

2 σ12 ≥ σ22 σ12 < σ22 1 F0 ≤ F1-α

3 σ12 ≤ σ22 σ12 > σ22 1 F0 ≥ Fα

EXAMPLE:

In comparing the variability of the tensile strength of two kinds of structural steel, an

experiment yielded the following results: n1 = 13, s12 = 19.2, n2 = 16 and s22 = 3.5, where

the units of measurements are thousand pounds per square inch. Assuming that the

measurements constitute independent random samples from two normal populations,

test the null hypothesis σ12 = σ22 against the alternative σ12 ≠ σ22 at α = 0.02 level of

significance.

The problem is solved using the eight-step hypothesis testing procedure as follows.

27

2. H0: σ12 = σ22

3. H1: σ12 ≠ σ22

4. α = 0.05

5. The test statistics is s12

F0 =

s 22

6. Reject H0 if F0 ≥ 2.96 (=Fα/2) or F0 ≤ 0.350 (=F1-α/2). Note that this results from

step 4, where we specified α = 0.02 (two tail).

Therefore, F0 = (19.2/3.5) = 5.49.

8. Conclusion: F0 = 5.49 > 2.96 (=Fα/2) the conclusion is to reject the null

hypothesis H0: σ12 = σ22 and conclude that the variability of the tensile strength

of the two kinds of steel is not the same.

i. when the alternative is of the type H1: σ12 > σ22 the conclusion would be to

reject H0: σ12 ≤ σ22at the 0.05 level of significance, since F0 = 5.49 > 2.40

(Fα) and conclude that the variability of the tensile strength of the first kind

of steel is greater than the variability of the tensile strength of the second

kind of steel.

ii. when the alternative is of the type H1: σ12 < σ22 the conclusion would be do

not reject H0: σ12 ≥ σ22at the 0.05 level of significance, since F0 = 5.49 >

0.371 (F1-α) and conclude that the variability of the tensile strength of the

first kind of steel is greater than the variability of the tensile strength of the

second kind of steel.

Chi-Square goodness of fit test is a non-parametric test that is used to find out how the

observed value of a given phenomena is significantly different from the expected value.

In Chi-Square goodness of fit test, the term goodness of fit is used to compare the

observed sample distribution with the expected probability distribution. Chi-Square

goodness of fit test determines how well theoretical distribution (such as normal,

binomial, or Poisson) fits the empirical distribution. In Chi-Square goodness of fit test,

sample data is divided into intervals. Then the numbers of points that fall into the

interval are compared, with the expected numbers of points in each interval.

28

PROCEDURE FOR CHI-SQUARE GOODNESS OF FIT TEST

assumes that there is no significant difference between the observed and the

expected value. In other words, the data follows a specified distribution.

b. Alternative hypothesis: In Chi-Square goodness of fit test, the alternative

hypothesis assumes that there is a significant difference between the

observed and the expected value. In other words, the data does not follow a

specified distribution.

2. Compute the value of Chi-Square goodness of fit test using the following formula:

k

(Oi − Ei ) 2

χ2 = ∑

i =1 Ei

where, χ = Chi-Square goodness of fit test statistic, O= observed value and

2

E= expected value.

freedom where k is the number of non-empty cells and c = the number of

parameters to be estimated + 1

depends on the distribution of the sample. The following table shows the

distribution and an associated degree of freedom:

Binominal distribution

2 n-2

(if p is estimated)

Poisson distribution 2 n-2

Normal distribution 3 n-3

same as in other tests, like Z- test, t-test, etc. The calculated value of Chi-Square

goodness of fit test is compared with the table value corresponding to (k-c)

degrees of freedom and at α level of significance. If the calculated value of Chi-

Square goodness of fit test is greater than the table value, we will reject the null

hypothesis and conclude that there is a significant difference between the

observed and the expected frequency. If the calculated value of Chi-Square

goodness of fit test is less than the table value, we will accept the null hypothesis

29

and conclude that there is no significant difference between the observed and

expected value.

EXAMPLES

1. For example, in 200 flips of a coin, one would expect 100 heads and 100 tails.

But what if 92 heads and 108 tails are observed? Would we reject the hypothesis

that the coin is fair? Or would we attribute the difference between observed and

expected frequencies to random fluctuation?

Solution:

of tails.

Alternative hypothesis: The frequency of heads is not equal to

the frequency of tails.

The expected frequencies in each of the two categories (heads or tails)

are not independent. To obtain the expected frequency of tails (100), we

need only to subtract the expected frequency of heads (100) from the

total frequency (200), or 200 - 100 = 100. Thus, given the expected

frequency in one of the categories, the expected frequency in the other is

readily determined. In other words, only the expected frequency in one of

the two categories is free to vary; that is, there is only 1 degree of

freedom.

Heads 92 100 -8 64 0.64

Tails 108 100 8 64 0.64

Total 200 200 0 χ2 = 1.28

Conclusion:

The critical values of χ2 for 1 degree of freedom, with α = .05 and α = .

01 are 3.841 and 6.635, respectively. As the calculated value of χ2 is less

than the table value at both α = .05 and α = .01 levels of significance we do

not reject the null hypothesis and conclude that the coin is fair. That is,

frequency of heads is equal to the frequency of tails.

test this hypothesis, we roll the die 300 times and observe the frequency

of occurrence of each of the faces. Because we hypothesized that the die

is unbiased, we expect that the number on each face will occur 50 times.

However, suppose we observe frequencies of occurrence as follows:

Face 1 2 3 4 5 6

30

Value

Occurrenc 42 55 38 57 64 44

e

difference to random fluctuation?

Solution:

occurrence of each of the six faces of the die is the same.

frequency of occurrence of each of the six faces of the die is not the

same.

There are six possible categories of outcomes: the occurrence of the six

faces. Under the assumption that the die is fair, we would expect that the

frequency of occurrence of each of the six faces of the die would be 50.

Note again that the expected frequencies in each of these categories are

not independent. Once the expected frequency for five of the categories

is known, the expected frequency of the sixth category is uniquely

determined, since the total frequency equals 300. Thus, only the

expected frequencies in five of the six categories are free to vary; there

are only 5 degrees of freedom associated with this example.

Face

O E O-E (O-E)2 (O-E)2/E

Value

1 42 50 -8 64 1.28

2 55 50 5 25 0.5

3 38 50 -12 144 2.88

4 57 50 7 49 0.98

5 64 50 14 196 3.92

6 44 50 -6 36 0.72

Total 300 300 0 10.28 = χ2

Conclusion:

The critical values of χ2 for 5 degree of freedom, with α = .05 and α = .

01 are 11.070 and 15.086, respectively. As the calculated value of χ2 is

less than the table value at both α = .05 and α = .01 levels of significance

we do not reject the null hypothesis and conclude that the die is fair. That

is, the frequency of occurrence of each of the six faces of the die is the

same.

31

3. The president of a major University hypothesizes that at least 90

percent of the teaching and research faculty will favor a new university

policy on consulting with private and public agencies within the state.

Thus, for a random sample of 200 faculty members, the president would

expect 0.90 x 200 = 180 to favor the new policy and 0.10 x 200 = 20 to

oppose it. Suppose, however, for this sample, 168 faculty members favor

the new policy and 32 oppose it. Is the difference between observed and

expected frequencies sufficient to reject the president's hypothesis that

90 percent would favor the policy? Or would the differences be attributed

to chance fluctuation?

Solution:

percent.

found by subtracting the expected number who supports it (180) from the

total number in the sample (200), or 200 - 180 = 20. Thus, given the

expected frequency in one of the categories, the expected frequency in

the other is readily determined. In other words, only the expected

frequency in one of the two categories is free to vary; that is, there is only

1 degree of freedom.

Favour 168 180 - 12 144 0.80

Oppose 32 20 12 144 7.20

Total 200 200 0 8.00 = χ2

Conclusion:

The critical values of χ2 for 1 degree of freedom, with α = .05 and α = .

01 are 3.841 and 6.635, respectively. As the calculated value of χ2 is

greater than the table value at both α = .05 and α = .01 levels of significance

we reject the null hypothesis and conclude that the faculty favouring the

new policy is not 90 percent.

32

CHI-SQUARE TEST FOR INDEPENDENCE OF ATTRIBUTES

CONTINGENCY TABLES

A frequency table in which a sample is classified according to the distinct classes of two

different attributes is called a contingency table. It is often of interest to test the

hypothesis that, in the population from which the sample was drawn, the two attributes

are independent. An mxn contingency table has m rows and n columns.

(Attribute 2) 1 2 ... j … n Total

1 O11 O12 O1j O1n R1

2 O21 O22 O2j O2n R2

. . . . .

i Oi1. Oi2. Oij Oin. Ri

. .

m Om1 Om2 Omj Omn Rm

Total C1 C2 Cj Cn N

The expected frequencies are obtained as follows: The expected frequency of the cell

corresponding to i-th row and j-th column is found by

i th RowTotal × j th ColumnTota l Ri × C j

Eij = =

GrandTotal N

Using the observed frequencies given in the contingency table and the expected

frequencies found using the above formula, we may test the null hypothesis that,

H0: The two attributes are independent (or) the two attributes are not associated.

H1: The two attributes are not independent (or) the two attributes are associated.

m

χ 2 = ∑∑

n (Oij − E ij ) 2

or simply for easy of understanding χ 2 = ∑

( O − E) 2 .

i j E ij E

The decision is to reject the null hypothesis H0 If the calculated value of χ2 , say χ0 is

2

degrees of freedom.

33

EXAMPLE

The following data were collected in a study on the effectiveness of inoculation for a

particular disease. The two attributes in this case are;

Attribute B

Attribute A Disease No disease

Inoculated 10 50

Not Inoculated 30 40

In this case the null hypothesis and alternative hypothesis are stated as,

Expected Frequencies

Inoculated 10 50 60

Not Inoculated 30 40 70

Total 40 90 130

If the null hypothesis is true then 40 people will contract the disease regardless of

whether or not they were inoculated. i.e., under the null hypothesis Ho, the probability of

contracting the disease is 40/130. Thus out of the 60 people inoculated we would

expect (40 x 60)/130 =18.5 of them to contract the disease. The expected frequencies

are therefore

Inoculated 18.5 41.5 60

Not Inoculated 21.5 48.5 70

Total 40 90 130

34

The test statistic is χ = ∑2 ( O − E) 2

E

χ =∑

2 ( O − E) 2 =

( 10 − 18.5)

2

+

( .)

50 − 415

2

+

( . )

30 − 215

2

+

( 40 − 48.5)

2

= 10.5

E 18.5 415

. 21.5 48.5

The critical value for a 1% significance level with 1 d.f. is 6.63. The null hypothesis is

therefore rejected at this level and it can be concluded that inoculation does have an

effect on the probability of contracting the disease. From the contingency table it can be

seen that inoculation reduces the risk.

Note

For chi-squared tests expected frequencies should be at least 5.

PROBLEMS

fatal accidents was dependent on the size of car involved. The following data were

collected:

Fatal 67 26 16

Non-fatal 128 63 46

Test the hypothesis that the probability of a fatality is dependent on the size of car.

2. Low birth weight in babies is defined as weights below 2500 grams. The following

table shows the number of low birth weight babies for three groups of mothers, non-

smokers, smokers and ex-smokers. Do the results suggest that the smoking habits

of the mother have an effect on birth weight?

Birth weight Non-smoker Smoker Ex-smoker

< 2500 grams 140 153 27

≥ 2500 grams 2197 1510 433

there is a difference in product acceptance by sex groups. In a market survey, 58

percent of 200 men questioned liked the product and 50 percent of 150 women

questioned liked the product. Is there a significant difference in product acceptance by

men and women?

35

2. The president of the college has reported that the average age of evening

students is 35 years. A random sample of 100 evening students was taken and it was

found that the average of the sample was 34 years with a SD of 5 years. At 1 % level

can we conclude that the resident’s claim is correct?

3. An educator claims that the average IQ of city college students is not more than

110. To test this claim, a random sample of 150 students was taken and gave relevant

test. Their average IQ score came to 11.2 with a standard deviation of 7.2. At level of

significance 0.05 test is the claim of the educator is justified.

20. Mean of the sample is 30. Test whether the sample has come from a population

with mean 29 at 5% level of significance.

5. A random sample of 400 men and 600 women were asked whether they would

like to have a flyover near their residence. 200 men and 325 were in favour of the

proposal. Test the hypothesis that proportion of men and women in favour of the

proposal are the same.

a factory conformed to specifications. An examination of a sample of 200 pieces of

equipment revealed that 18 were faulty. Test his claim at a significance level of 5%

and 1%.

7. A die was thrown 9000 times and through of 3 or 4 observed 3240 times. Show

that the die can not be regarded as unbiased one.

8. In a sample of 600 men from a certain large city, 450 are found be smokers. In

one of 900 from another large city 450 are smokers. Do the data indicate that the

cities are significantly different with respect to the prevalence of smoking among men?

9. In a year there are 956 births in a town A of which 52.5% were male while in

town A and B combined this proportion in a total of 1406 birth was 0.496. Is there any

significance difference in the proportion of male births in the two towns?

10. In two large populations, there are 30 and 25% respectively for fair haired

people. Is this difference likely to be hidden in samples of 1200 and 900 respectively

from the two populations?

11. A machine puts out 16 imperfect articles in a sample of 500. After machine is

overhauled, it puts out 3 imperfect articles in a batch of 100. Has the machine

improved?

centers provide the examination scores in the following table:

36

Examination Score Results

Training centre Training Centre

A B

Sample size 1. 30 2. 40

Mean 3. 82 4. 78

SD 5. 8 6. 10

Using the above data test whether the two recent centers differ in terms of

educational quality

13. Suppose that 100 tires made by a certain manufacturer lasted on the average

21819 miles with a standard deviation of 1295 miles. Test the null hypothesis that

µ=22000 miles against the alternative hypothesis µ≠22000, at the 0.05 level of

significance.

37

- Hypothesis TestingTransféré parDivyamathi Thayumanavar
- Chapter 2 Hypothesis TestingTransféré parMario M. Ramon
- QTMTransféré parAishwarya A.G
- Null Hypothesis, Statistics, Z-Test, SignificanceTransféré parSherwan R Shal
- Statistical Significance HypothesisTransféré parSherwan R Shal
- Steps in Hypothesis TestingTransféré parIzi
- QuantitativeTechniquesInManagement-MB007-Question.docxTransféré parAiDLo
- Hypothesis Testing NotesTransféré parBu-Hilal Alsuwaidi
- BusinessLaw MB012 QuestionTransféré parAiDLo
- 7. Confidence Intervals VIEWTransféré parRahul Sathyamoorthy
- Z-testTransféré parAmrit Keyal
- ADL 07 Quantitative Techniques in Management v4Transféré parEDUCARE INSTITUTE
- US-64Transféré parrocks777
- ADL 07 Quantitative Techniques in Management AM4Transféré parsolvedcare
- Assignement - QTM(Quantitive Techique Mgt)Transféré parNimit Gupta
- Student's_t-testTransféré parSanjay Lakhtariya
- Practice Exercises for Test of HypothesisTransféré parNicole Mangibin
- UTI-DevTransféré parSamir Sharma
- Multiple Choice QuestionesTransféré parDamic
- ADL-07-Ver1+Transféré parSanjeev Kumar
- Case US-64 ControversyTransféré parBiwesh
- Amity Pan African MBA BBA Solved AssignmentsTransféré parAlhassan Tejan
- Managerial EconomicsTransféré parmssharma5004
- Test of HypothesisTransféré parJyoti Prasad Sahu
- AMBO205_Business Transformation AssignmentTransféré parAnant Bhargava
- Research MethodologyTransféré parjaminkwad
- AssigmentTransféré parAmori Geofrey
- Critical evaluation of Baumol modelTransféré parWaleedSami
- ADL-12 AssignmentTransféré parRavi Prasad Pedapati
- Assignment 1 SolTransféré parkshambl

- Ai50Proceedings WebTransféré parConstanzaPérez
- Practical Research 2Transféré parCarl Correos
- Early Childhood AustraliaTransféré parCatherine TandooriGalaxy Abbott
- Western philosophy is the love of wisdom and Islamic philosophy is wisdom it selfTransféré parBaytoona
- GusfieldTransféré parBelen Capitan
- 1Translation and MedicineTransféré parMN
- ceng4351.pdfTransféré paryoseph dejene
- assesment qualitatifTransféré parSalahuddin Ageng Natanegara
- (eBook - IsLAM) - Life After DeathTransféré paranon-982833
- Geoffrey Galt Harpham - Roots, Races and the Return to PhilologyTransféré parLucas Santos
- Polanyi 1967Transféré parTirso Medellín Barquín
- Yoga TherapyTransféré parIsmet Korac
- The New Feudalists : The EU Communitarian Agenda.Transféré parPhilip John Jones
- Passerini Luisa 2009 the Problematic Intellectual Repercussions of 68Transféré parMariana Perry
- Research MethodologyTransféré parKavita Poddar
- (10) Marwa Knowledge in MotionTransféré parAmr Akkad
- 828Transféré parAbdulla Jawad Alshemary
- Introduction from Meat, Medicine and Human Health in the Twentieth CenturyTransféré parPickering and Chatto
- GIS Methodologies for Developing Conservation Strategies-gPGTransféré parAravind Kanth
- Legal Realist PerspectiveTransféré parJoshua Paulo Ocampo
- BinaTransféré parmj riego
- Quantitative Techniques IntroductionTransféré parexotic_ap
- Guide to Portfolios and ExhibitionsTransféré parshaileshvc
- C. Robert Cloninger, Feeling Good - the Science of Well-BeingTransféré parIovana Gherman
- ttTransféré parNibha Pandey
- Four Basic Questions About Artificial Intelligence – Pei WangTransféré parTom Yates
- Political Law Review - UST Golden NotesTransféré parAnonymous 33LIOv6L
- Jayatilleke - Early Buddhist Theories of Knowledge.pdfTransféré paryanez de gomera
- LP in Science and Health III 2nd RatingTransféré parRox Francisco
- Scientific Method Lab ReportTransféré parRick Crackpipe Zahn