Statistical Inferences about Two Populations

Learning Objectives
Test hypotheses and construct confidence intervals about the difference in two population means using the Z statistic. Test hypotheses and construct confidence intervals about the difference in two population means using the t statistic.

QT-II/Ses 3/SAMPRIT

Learning Objectives
Test hypotheses and construct confidence intervals about the differences in two population proportions. Test hypotheses and construct confidence intervals about two population variances.

QT-II/Ses 3/SAMPRIT

Sampling Distribution of the Difference Between Two Sample Means

x X1

x x

x
1

x x X1 X2
1

x X X1 x2
1

x
2

x X2
Population 2

QT-II/Ses 3/SAMPRIT

Sampling Distribution of the Difference between Two Sample Means

x1 1 xX 2 X 2

11

2 2

x1 x2

X1 X 2

n11 n

2 2 11

n n2
2

2 2 2

X xX 1 x2 2 1

X x x1 X 2
1 2

QT-II/Ses 3/SAMPRIT

Z Formula for the Difference in Two Sample Means

When 12 and 22 are known and Independent Samples

x x
1

2 2 1

1 2 2 2

n n
1
QT-II/Ses 3/SAMPRIT

Hypothesis Testing for Differences Between Means: Example with CI 95%

Rejection Region

.025
2

.025

Non Rejection Region

x x X1 X2
1

Critical Values
QT-II/Ses 3/SAMPRIT

Hypothesis Testing for Differences Between Means: The Wage Example (part 2)
Rejection Region

Rejection Region

If z < - 1.96 or z > 1.96, reject Ho. If - 1.96 z 1.96, do not reject Ho.

.025
2
Non Rejection Region
1.96 1.96

.025

zc c Z

0 Critical Values

zc c 1.96 Z 1 96

QT-II/Ses 3/SAMPRIT

Example
We want to conduct a Hypothesis test to determine whether the average annual wage for an advertising manager is different from the average annual wage of an auditing manager. A random sample of 32 advertising managers from across US is taken. The advertising managers are contacted by telephone and asked what is their annual salary. A similar random sample is taken of 34 auditing managers. The resulting salary data are given in the next table, along with the sample means, the sample standard deviations and the sample variances.
QT-II/Ses 3/SAMPRIT

Crucial points
The analyst is testing whether there is any difference in the average wage of an advertising manager and an auditing manager So it is a two tailed test If testing is on whether one was paid more than the other, the test would have been one tailed

QT-II/Ses 3/SAMPRIT

Example
Advertising Managers
74.256 57.791 71.115

Auditing Managers

96.234
89.807 93.261 103.030 74.195 75.932 80.742

65.145
96.767 77.242 67.056 64.276 74.194 65.360

67.574
59.621 62.483 69.319 35.394 86.741 57.351

n x

1 1 1 2 1

32 70.700 16.253 264.164

69.962 55.052 57.828 63.362 37.194 99.198 61.254

77.136 66.035 54.335 42.494 83.849 67.160 37.386 59.505 72.790 71.351 58.653 63.508

43.649 63.369 59.676 54.449 46.394 71.804 72.401 56.470 67.814 71.492

39.672
45.652 93.083 63.384

73.904
54.270 59.045 68.508

n x

2 2 2 2 2

34 62.187 12.900 166.411

73.065 48.036 60.053 66.359 61.261

QT-II/Ses 3/SAMPRIT

Working
If zZ < -1.96 or Z > 1.96,, reject Ho.0 . If 1.96 z 1.96 reject H If - 1.96 Z 1.96,,do not reject HH 0 . If 1.96 z 1.96 do not reject o.

Rejection Region

Rejection Region

zZ
2 .025
Non Rejection Region

( x1 1 x2X 2 ( X )
2 2 1 1
1

.025

S S n1n nn 2

1 1 22 22
2

2) 2

z Zc

21.96 .33

0 Critical Values

2 33 zZ c 1..96 c

(70.700 - 62.187 - (0) 70.700 62.187) 0 2.2.35 35 256.253 166.411 166.411 256 32 34 32 34
Since z =2.35 > 1.96,, reject Ho.0 . Since Z 2.35 1.96 reject H

QT-II/Ses 3/SAMPRIT

Example
A sample of 87 working professional working women showed that the average amount paid annually into a private pension fund per person was $3,352, with a sample standard deviation of $1,100. A sample of 76 professional working men showed that the average amount paid annually into a private pension fund per person was $5,727, with a sample standard deviation of $1,700. A womens activist group wants to prove that women do not pay as much per year as men into private pension funds. If they use CI = 99% and the sample data, will they be able to reject a null hypothesis that women annually pay the same as or more than men into private pension funds?
QT-II/Ses 3/SAMPRIT

Crucial points
Test is one tailed When the population variances are unknown and the sample sizes are large (n1 and n2 greater than equal to 30), sample variances can be used For large samples, sample variances are good approximations of population variances

QT-II/Ses 3/SAMPRIT

Working

Ho: Ha:

0 0
zc c Z

Rejection Region

.001
Non Rejection Region
3 3..08

Critical Value

QT-II/Ses 3/SAMPRIT

Working
Women x1 $3,352

Men x2 $5,727
2

Rejection Region
.001

$1,100

$1,700 76

n1 87

n2

Non Rejection Region

x x
1

2 2 1

1 2 2 2

33.08 .08

0 Critical Value

n n
1

3352 5727 1100 87

0
2

10.42

If z < - 3.08, reject Ho. If z 3.08, do not reject Ho.

1700 76

Since z = - 10.42 < - 3.08, reject Ho.

Confidence Interval to Estimate are known

When

1,

2 1 2

2 2

x x
1

x x
1

QT-II/Ses 3/SAMPRIT

Example
A consumer test group wants to determine the difference in gasoline mileage of cars using regular unleaded gas and cars using premium unleaded gas. Researchers for the group divided a fleet of 100 cars of the same make in half and tested each car on one tank of gas. Fifty of the cars were filled with regular unleaded gas and 50 were filled with premium unleaded gas. The sample average for the regular gasoline group was 21.45 miles per gallon, with a standard deviation of 3.46 mpg. The sample average for the premium gasoline group was 24.6 mpg, with a standard deviation of 2.99 mpg. Construct a 95% confidence interval to estimate the difference in the mean gas mileage between the cars using regular gasoline and the cars using premium gasoline

QT-II/Ses 3/SAMPRIT

Working
Re gular
Pr emium

n x

1 1 1

50 21.45 3.46

n x

2 2 2

50 24.6 2.99
2 2 2 1 2 2 2 1 2 2 2 2 2 2

95% Confidence

z = 1.96

x x
1

21.45 24.6 1.96

n n 3.46 2.99
1 2

x x
1

n n
1

50

50 4.42

3.46 21.45 24.6 1.96 50 1.88

2.992 50

QT-II/Ses 3/SAMPRIT

The t Test for Differences in Population Means

QT-II/Ses 3/SAMPRIT

t Formula to Test the Difference in Means Assuming 12 = 22

( x1

x2 ) (

) 1 n1 1 n2

2 2 s1 (n1 1) s2 (n2 n1 n2 2

1)

QT-II/Ses 3/SAMPRIT

Example
At the Hernandez Manufacturing Company, an application of the test of the difference in small sample mean arises. New employees are expected to attend a three-day seminar to learn about the company. At the end of the seminar, they are tested to measure their knowledge about the company. The traditional training method has been lecture and a QnA session. Management decided to experiment with a different training procedure, which processes new employees in two days by using vcd and having no QnA session. If, this procedure works, it could save the company thousands of dollars over a period of several years. However, there is some concern about the effectiveness of the two day method, and the company managers would like to know whether there is any difference in the effectiveness of the two training methods. To test the difference in the two methods, the managers randomly select one group of 15 newly hired employees to take the three day seminar (method A) and a second group of 12 new employees for the two day vcd method (method B)
QT-II/Ses 3/SAMPRIT

Example
The table show the test scores of the two groups. Using = 0.05, the managers want to determine whether there is a significant difference in the mean scores of the two groups. They assume that the scores for this test are normally distributed and that the population variances are approximately equal.
Training Method A
56 47 42 50 51 52 53 42 45 43 52 48

Training Method B
59 52 53 54 57 56 55 64 53 65 53 57

47

44

44

QT-II/Ses 3/SAMPRIT

Solution
Ho: Ha:
2 df

0 0
Rejection Region Rejection Region

.05 .025 2 n1 n2 2 15 12 2 2.060

.025

25
t

.025
Non Rejection Region

t0.25, 25

.025, 25

2.060

0 Critical Values

.025, 25

2.060

If t < - 2.060 or t > 2.060, reject Ho. If - 2.060 t 2.060, do not reject Ho.
QT-II/Ses 3/SAMPRIT

Solution
Training Method A
56 47 42 50 47 51 52 53 42 44 45 43 52 48 44

Training Method B
59 52 53 54 57 56 55 64 53 65 53 57

n1 15 x1 s12 47.73 19.495

n2 x2
2 s2

12 56.5 18.273

QT-II/Ses 3/SAMPRIT

Solution
t ( x1 x2 ) (
1 2

) 1 n1 1 n2 1 12

2 2 s1 (n1 1) s2 (n2 n1 n2 2

1)

47.73 56.50 0 19.495 14 18.273 11 1 15 12 2 15 5.20

If t < - 2.060 or t > 2.060, reject Ho. If - 2.060 t 2.060, do not reject Ho.

Since t = -5.20 < -2.060, reject Ho.

QT-II/Ses 3/SAMPRIT

Example
Is there any difference in the way Chinese cultural values affect the purchasing strategies of industrial buyers in Taiwan and mainland China? A study by researchers at the National university in Taiwan attempted to determine whether there is a significant difference in the purchasing strategies of industrial buyers in the two countries based on the cultural dimension labeled integration. Integration is being in harmony with ones self, family, and associates. For the study, 46 Taiwanese buyers and 26 mainland Chinese buyers were contacted and interviewed. Buyers were asked to respond to 35 using a 9-point scale with possible answers ranging from no importance (1) to extreme importance (9). The resulting statistics for the two groups are shown in the table. Using = 0.01, test to determine whether there is a significant difference between buyers of the two countries on integration.

QT-II/Ses 3/SAMPRIT

Example
Taiwanese Buyers Sample size is 46 Mean is 5.42 Sample Variance is 0.3364 Mainland Chinese Buyers Sample size is 26 Mean is 5.04 Sample Variance is 0.2401

QT-II/Ses 3/SAMPRIT

Confidence Interval to Estimate 2 and 2 are unknown and 1 2

1 1

when
2 2

( x1

s (n1 1) s (n2 1) 1 x2 ) t n1 n2 2 n1 n1 n2 2

1 n2

where df

QT-II/Ses 3/SAMPRIT

Example
Construct a 99% Confidence Interval from the following data
Sample Size = 9 Mean = 37.09 Sample sd = 1.727 t for 0.005 and 17 is 2.898 Sample Size = 10 Mean = 34.99 Sample sd = 1.253

QT-II/Ses 3/SAMPRIT