1. Hypothesis Tests for One Population Mean
Chapter 9
Hypothesis Tests for
One Population Mean

2. Definition 9.1 This is just a true or false; is or isn’t…
Null: Mu = mu0
Alt: Mu not= mu0 2 tailed
Mu < mu0 Left tailed
Mu > mu0 Right tailed

3. Example 9.4
A company that produces snack foods uses a machine to package 454 g bags of pretzels. We assume that the net weights are normally distributed and that the population standard deviation of all such weights is 7.8 g.  A simple random sample of 25 bags of pretzels has the net weights, in grams, displayed in Table 9.1. Do the data provide sufficient evidence to conclude that the packaging machine is not working properly? We use the following steps to answer the question.
Insert table 6.1 from page 276
Table 9.1

4. a. State the null and alternative hypotheses for the hypothesis test.
Example 9.4
a. State the null and alternative hypotheses for the hypothesis test.
b. Discuss the logic of this hypothesis test.
c. Identify the distribution of the variable , that is, the sampling distribution of the sample mean for samples of size 25.
d. Obtain a precise criterion for deciding whether to reject the null hypothesis in favor of the alternative hypothesis.
e. Apply the criterion in part (d) to the sample data and state the conclusion.
Insert table 6.1 from page 276

5. a. The null and alternative hypotheses, as stated in Example 9.1, are
Solution Example 9.4
a. The null and alternative hypotheses, as stated in Example 9.1, are
H0 : μ = 454 g (the packaging machine is working properly)
Ha : (the packaging machine is not working properly).
If the null hypothesis is true, then the mean weight,
, of the sample of 25 bags of pretzels should approximately equal 454 g. However, if the sample mean weight differs “too much” from 454 g, we would be inclined to reject the null hypothesis and conclude that the alternative hypothesis is true. As we show in part (d), we can use our knowledge of the sampling distribution of the sample mean to decide how much difference is “too much.”
Insert table 6.1 from page 276

6. c. n = 25, σ = 7.8, weights are normally distributed,
Solution Example 9.4
c. n = 25, σ = 7.8, weights are normally distributed,
･ μ = μ (which we don’t know),
･ σ = and
･ is normally distributed.
In other words, for samples of size 25, the variable is normally distributed with mean μ and standard deviation 1.56 g.
d. The “95.44” part of the 68.26–95.44–99.74 rule states that, for a normally distributed variable, 95.44% of all possible observations lie within two standard deviations to either side of the mean. Applying this part of the rule to the variable and referring to part (c), we see that 95.44% of all samples of 25 bags of pretzels have mean weights within 2•1.56 = 3.12 g of μ.
Insert table 6.1 from page 276

7. Solution Example 9.4
d. Equivalently, only 4.56% of all samples of 25 bags of pretzels have mean weights that are not within 3.12 g of μ, as illustrated in Fig. 9.1.
Insert table 6.1 from page 276
Figure 9.1

8. Solution Example 9.4
d. If the mean weight, , of the 25 bags of pretzels sampled is more than two standard deviations (3.12 g) from 454 g, reject the null hypothesis and conclude that the alternative hypothesis is true. Otherwise, do not reject the null hypothesis.
2 SD is not a standard but is widely used.
Figure 9.2

9. Solution Example 9.4
The mean weight, , of the sample of 25 bags of pretzels whose weights are given in Table 9.1 is 450g. So, z =( − 454) / 1.56 = (450 − 454) /1.56 = −2.56.
That is, the sample mean of 450 g is 2.56 standard deviations below the null-hypothesis population mean of 454 g, as shown in Fig. 9.3.
Z is the number of SDs off the avg. Explain why.
Figure 9.3

10. Solution Example 9.4
Because the mean weight of the 25 bags of pretzels sampled is more than two standard deviations from 454 g, we reject the null hypothesis (μ = 454 g) and conclude that the alternative hypothesis ( ) is true.
The data provide sufficient evidence to conclude that the packaging machine is not working properly.
Insert table 6.1 from page 276

11. The set of values for the test statistic that leads us to reject the null hypothesis is called the rejection region. In this case, the rejection region consists of all z-scores that lie either to the left of −2 or to the right of 2; that part of the horizontal axis under the shaded areas in Fig. 9.5.
This is 2 sigma left or right.
Need to know:
How to choose null and alternate
How to classify as two, right, or left tailed
Logic behind hypothesis test.
Figure 9.5

12. Definition 9.2
Test statistic is z from previous slide.

13. For a two-tailed test, the null hypothesis is rejected when the test statistic is either too small or too large. The rejection region consists of two parts: one on the left and one on the right, (Fig. 9.6(a)).
For a left-tailed test, the null hypothesis is rejected only when the test statistic is too small. The rejection region consists of only one part, on the left, (Fig.9.6(b)).
For a right-tailed test, the null hypothesis is rejected only when the test statistic is too large. The rejection region consists of only one part, on the right, (Fig.9.6(c)).
Change to page 407
Figure 9.6

14. Definition 9.3
Use previous example to illustrate these errors

15. Definition 9.4 These are the tails our distribution
1 - confidence level.

16. Key Fact 9.1 Beta covered later in book, but not by us.
Small alpha means large beta.

17. Key Fact 9.2 This is because a small alpha means a large beta.
This is a non-committal answer.
End of Day 6

18. Key Fact 9.3
Insert Definition 9.3 from page 408

19. Procedure 9.1
Z is how many SDs mu is away from the avg.

20. Procedure 9.1 (cont.)
Change to page 417

21. Key Fact 9.4 This should look familiar Do example 7, 8, and or 9
HW 9.54, 55, 57, 59, 62

22. H0 : μ = 26 mpg (manufacturer’s claim)
Example 9.10
The manufacturer of a new model car, the Orion, claims that a typical car gets 26 miles per gallon (mpg). A consumer advocacy group is skeptical of this claim and thinks that the mean gas mileage, μ, of all Orions may be less than 26 mpg. The group plans to perform the hypothesis test
H0 : μ = 26 mpg (manufacturer’s claim)
Ha : μ < 26 mpg (consumer group’s conjecture),
at the 5% significance level, using a sample of 30 Orions. Do the data provide sufficient evidence that the manufacturer’s claim is incorrect for:
a mpg. b mpg.
Assume that gas mileages of Orions are normally distributed with a standard deviation of 1.4 mpg.
Insert table 6.1 from page 276

23. The decision criterion can thus be expressed in terms of
Solution Example 9.10
The inference under consideration is a left-tailed hypothesis test for a population mean at the 5% significance level. The test statistic is
and the critical value is −zα = −z0 .05 = − Thus the decision criterion for the hypothesis test is: If z ≤ −1.645, reject H0; if z > −1.645, do not reject H0. Computing error probabilities is somewhat simpler if the decision criterion is expressed in terms of instead of z.
The decision criterion can thus be expressed in terms of
as: If ≤ 25.6 mpg, reject H0; if > 25.6 mpg, do not reject H0. See Fig
Insert table 6.1 from page 276

24. Solution Example 9.10
Insert table 6.1 from page 276
Figure 9.14

25. is normally distributed.
Solution Example 9.10
a. If μ = 25.8 mpg, then
and
is normally distributed.
Thus, the variable is normally distributed with a mean of 25.8 mpg and a standard deviation of 0.26 mpg. The normal curve for is shown in Fig
Insert table 6.1 from page 276

26. Change to page 426
Figure 9.15

27. A Type II error occurs if we do not reject H0, that is, if
Solution Example 9.10
A Type II error occurs if we do not reject H0, that is, if
> 25.6 mpg. The probability of this happening equals the percentage of all samples whose means exceed 25.6 mpg, which we obtain in Fig Thus, if the true mean gas mileage of all Orions is 25.8 mpg, the probability of making a Type II error is ; that is, β =
Interpretation There is roughly a 78% chance that the consumer group will fail to reject the manufacturer’s claim that the mean gas mileage of all Orions is 26 mpg when in fact the true mean is 25.8 mpg.
Although this result is a rather high chance of error, we probably would not expect the hypothesis test to detect such a small difference in mean gas mileage (25.8 mpg as opposed to 26 mpg) with a sample size of only 30.
Insert table 6.1 from page 276

28. Solution Example 9.10
b. We proceed as we did in part (a) but this time assume that μ = 25.0 mpg. Fig shows the required computations.
Insert table 6.1 from page 276
Figure 9.16

29. Solution Example 9.10
If the true mean gas mileage of all Orions is 25.0 mpg, the probability of making a Type II error is ; that is, β =
Interpretation There is only about a 1% chance that the consumer group will fail to reject the manufacturer’s claim that the mean gas mileage of all Orions is 26 mpg when in fact the true mean is 25.0 mpg.
Insert table 6.1 from page 276

30. We combined Figs. 9. 15 and 9. 16 with two others. Figure 9
We combined Figs and 9.16 with two others. Figure 9.17 shows clearly that the farther the true mean is from the null hypothesis mean of 26 mpg, the smaller will be the probability of a Type II error. This result is hardly surprising: We would expect that a false null hypothesis is more likely to be detected when the true mean is far from the null hypothesis mean than when the true mean is close to the null hypothesis mean.
Change to page 428
Figure 9.17

31. Definition 9.5
Insert Definition 9.5 from page 428

32. In reality, the true value of the parameter in question will be unknown. Consequently, constructing a table of powers for various values of the parameter is helpful in evaluating the effectiveness of the hypothesis test.
For the gas mileage illustration – where the parameter in question is the mean gas mileage, μ, of all Orions – we have already obtained the Type II error probability, β, when the true mean is 25.8 mpg, 25.6 mpg, 25.3 mpg, and 25.0 mpg, as depicted in Fig
Similar calculations yield the other β probabilities shown in the second column of Table 9.8. The third column of Table 9.8 shows the power that corresponds to each value of μ, obtained by subtracting β from 1.
Change to page 429

33. Change to page 429
Table 9.8

34. We can use Table 9.8 to evaluate the overall effectiveness of the hypothesis test. We can also obtain from Table 9.8 a visual display of that effectiveness by plotting points of power against μ and then connecting the points with a smooth curve. The resulting curve is called a power curve and is shown in Fig on the next slide. In general, the closer a power curve is to 1 (i.e., the horizontal line 1 unit above the horizontal axis), the better the hypothesis test is at detecting a false null hypothesis.
Change to page 429

35. Change to page 431
Figure 9.18

36. The smaller we specify the significance level, the smaller will be the power. However, by using a large sample, we can have both a small significance level and large power. Here is a graph of power for  = 0.05 and samples sizes of n = 30 and n = 100.
Figure 9.20
Change to page 431

37. Key Fact 9.5
Insert Definition 9.5 from page 428

38. Definition 9.6
Change to page 434

39. If the null hypothesis is true, this test statistic has the standard normal distribution, and its probabilities equal areas under the standard normal curve. If we let z0 denote the observed value of the test statistic z, we obtain the P-value as follows:
Two-tailed test: The P-value is the probability of observing a value of the test statistic z that is at least as large in magnitude as the value actually observed, which is the area under the standard normal curve that lies outside the interval from − |z0 | to |z0 |, as in Fig. 9.21(a).
Change to page 435
Figure 9.21

40. Left-tailed test: The P-value is the probability of observing a value of the test statistic z that is as small as or smaller than the value actually observed, which is the area under the standard normal curve that lies to the left of z0, as in Fig. 9.21(b).
Right-tailed test: The P-value is the probability of observing a value of the test statistic z that is as large as or larger than the value actually observed, which is the area under the standard normal curve that lies to the right of z0, as in Fig. 9.21(c).
Figure 9.21
Change to page 435

41. The P-value can be interpreted as the observed significance level of a hypothesis test. Suppose that the value of the test statistic for a right-tailed z-test turns out to be Then the P-value of the hypothesis test is 0.03 (actually ), the shaded area in Fig The null
hypothesis would be rejected for a test at the 0.05 significance level but would not be rejected for a test at the 0.01 significance level. In fact, the P-value is precisely the smallest significance level at which the null hypothesis would be rejected.
Change to page 437
Figure 9.24

42. Procedure 9.2
Change to page 438

43. Procedure 9.2 (cont.)
Change to page 438

44. Table 9.11
HW 9.117, 120

45. Change to page 440
Table 9.12

46. Use Table IV to estimate the P-value of each one-mean t-test.
Example 9.16
Use Table IV to estimate the P-value of each one-mean t-test.
Left-tailed test, n = 12, and t = −1.938
b. Two-tailed test, n = 25, and t = −0.895
Solution a. Because the test is left-tailed, the P-value is the area under the t-curve with df = 12−1 = 11 that lies to the left of −1.938, as shown in Fig. 9.27(a).
Same as z test but we do not know sigma
Figure 9.27

47. Solution Example 9.16
A t-curve is symmetric about 0, so the area to the left of −1.938 equals the area to the right of 1.938, which we can estimate by using Table IV. In the df = 11 row of Table IV, the two t-values that straddle are
t0.05 = and t0.025 = Therefore the area under the t-curve that lies to the right of is between and 0.05, as shown in Fig. 9.27(b). Consequently, the area under the t-curve that lies to the left of −1.938 is also between and 0.05, so < P < Hence we can reject H0 at any significance level of 0.05 or larger, and we cannot reject H0 at any significance level of or smaller. For significance levels between and 0.05, Table IV is not sufficiently detailed to help us to decide whether to reject H0. 
Sigma unknown we use t test vs z test
Everything else the same.

48. Solution Example 9.16
b. Because the test is two tailed, the P-value is the area under the t-curve with df = 25 − 1 = 24 that lies either to the left of −0.895 or to the right of 0.895, as shown in Fig. 9.28(a).
Change to page 447
Figure 9.28

49. df = 24 row of Table IV, 0.895 is smaller than any other
Solution Example 9.16
Because a t-curve is symmetric about 0, the areas to the left of −0.895 and to the right of are equal. In the
df = 24 row of Table IV, is smaller than any other
t-value, the smallest being t0.10 = The area under the t-curve that lies to the right of 0.895, therefore, is greater than 0.10, as shown in Fig. 9.28(b). Consequently, the area under the t-curve that lies either to the left of −0.895 or to the right of is greater than 0.20, so P > Hence we cannot reject H0 at any significance level of 0.20 or smaller. For significance levels larger than 0.20, Table IV is not sufficiently detailed to help us to decide whether to reject H0.
Change to page 447

50. Procedure 9.3
Change to page 448, Remove Critical Value Approach See book.

51. Procedure 9.3 (cont.)
Example 9.17 if time
HW 9.141, 142, 143

52. Example 9.19
The U.S. Department of Agriculture publishes information about food costs in Agricultural Research Service. According to that document, a typical U.S. family of four spends about $157 per week on food. Ten randomly selected Kansas families of four have the weekly food costs shown in Table Do the data provide sufficient evidence to conclude that the mean weekly food cost for Kansas families of four is less than the national mean of $157?
Change to page 447
Table 9.14

53. Solution Example 9.19
Let μ denote the mean weekly food cost for all Kansas families of four. We want to perform the hypothesis test H0 : μ = $157 (mean weekly food cost is not less than $157) Ha : μ < $157 (mean weekly food cost is less than $157). As we said, a condition for the use of the Wilcoxon signed-rank test is that the variable under consideration have a symmetric distribution. If the weekly food costs for Kansas families of four have a symmetric distribution, a graphic of the sample data should be roughly symmetric. A stem-and-leaf diagram of the sample data is roughly symmetric and so does not reveal any obvious violations of the symmetry condition.  We therefore apply the Wilcoxon signed-rank test to carry out the hypothesis test.
Change to page 447

54. μ0 = $157. The steps for doing so are presented in Table 9.15.
Solution Example 9.19
To begin, we rank the data in Table 9.14 according to distance and direction from the null hypothesis mean,
μ0 = $157. The steps for doing so are presented in Table 9.15.
Change to page 456
Table 9.15

55. Solution Example 9.19
The absolute differences, |D|, displayed in the third column, identify how far each observation is from 157. The ranks of those absolute differences, displayed in the fourth column, show which observations are closer to 157 and which are farther away. The signed ranks, R, displayed in the last column, indicate additionally whether an observation is greater than 157(+) or less than 157(−). Figure 9.32 depicts the information for the second and third rows of Table 9.15.
Change to page 447

56. Change to page 447
Figure 9.32

57. Procedure 9.4
Delete Slide, Insert Procedure 9.4 page 459

58. Procedure 9.4 (cont.)
Delete slide and update with continuation on page 459

59. Flowchart for choosing the correct hypothesis testing procedure for a population mean.
Change to page 468
Figure 9.34

60. Homework
When able, do the following by hand and check your answer with minitab.
9.54, 55, 57, 59, 62
9.117, 120
9.141, 142, 143
Example 9.17 if time
HW 9.141, 142, 143