2. Enter the following 25 data values beginning in cell A11 and continue down the A column to cell A35:

42, 30, 53, 50, 52, 30, 55, 49, 61, 74, 26, 58, 40,

40, 28, 36, 30, 33, 31, 37, 32, 37, 30, 32, 23

3. Select the following cells and type the headings indicated below:

In cell A10 type __Sales__

In cell D11 type __Mean__

In cell D12 type __Standard Deviation__

In cell D13 type __# Observations__

In cell D15 type __Confidence Interval for Alpha = .10__

In cells D16, D19, and D22 type __Min.__

In cell D18 type __Confidence Interval for Alpha = .05__

In cell D21 type __Confidence Interval for Alpha = .01__

In cells F16, F19, and F22 type __Max.__

4. Select cell **C11**. Then click on the **Formulas tab** in the toolbar, then select **More Functions**. Under the Function category select **Statistical.** Under the Function name select **AVERAGE**. In the dialog box Number 1 type **A11:A35** then click on **OK**. The mean of the 25 data values should appear.

5. Select cell **C12**. Then click on the **Formulas tab** in the toolbar, then select **More Functions**. Under the Function category select **Statistical.** Under the Function name select **STDEV.P**. In the dialog box Number 1 type **A11:A35** then click on **OK**. The standard deviation of the 25 data values should appear.

6. Select cell **C13**. Then click on the **Formulas tab** in the toolbar, then select **More Functions**. Under the Function category select **Statistical.** Under the Function name select **COUNT**. In the dialog box Value 1 type **A11:A35** then click on **OK**. The number of observations (25) for the data values should appear.

7. Select cell **C15**. Then click on the **Formulas tab** in the toolbar, then select **More Functions**. Under the Function category select **Statistical.** Under the Function name select **CONFIDENCE.NORM** (or CONFIDENCE for Excel 2007). In the dialog box Alpha type **.10**, in the dialog box Standard deviation type **C12**, in the dialog box Size type **C13**. Then click on **OK**. The confidence interval (± value) should appear.

(Instructions continue on the next page)

8. Select cell **C18**. Repeat Step 7 above from the second sentence, except type **.05** in the Alpha dialog box. Then click **OK**. The confidence interval (± value) should appear.

9. Select cell **C21**. Repeat Step 7 above from the second sentence, except type **.01** in the Alpha dialog box. Then click **OK**. The confidence interval (± value) should appear.

10. Select cell **C16** and type **=C11-C15**. Then hit Enter. Next select cell **E16** and type **=C11+C15**, and hit Enter. The minimum and maximum values for the confidence interval should appear.

11. Select cell **C19** and type **=C11-C18**. Then hit Enter. Next select cell **E19** and type **=C11+C18**, and hit Enter. The minimum and maximum values for the confidence interval should appear.

12. Select cell **C22** and type **=C11-C21**. Then hit Enter. Next select cell **E22** and type **=C11+C21**, and hit Enter. The minimum and maximum values for the confidence interval should appear.

13. Save the worksheet on a disk as CONFIDENCE.xlsx and print-out the worksheet to submit to the instructor.

14. In addition to submitting a print-out of the worksheet, also reference the data in your print-out and your Notes.

(a) Draw three separate graphs (one for each alpha value .10, .05, and .01) illustrating the alpha values, confidence levels, the mean, and the minimum and maximum values for the confidence level.

(b) What is the Maximum Tolerable Error value at an Alpha = .10, Alpha = .05, and an Alpha = .01? answer next to each graph.

(c) Now, suppose the data represents a population and you are performing hypothesis tests.

(1) Indicate and show on the graphs whether you would accept or reject

the H_{0} for each of the three alpha values (.10, .05, .01) if **x-bar = **

**45.9** and the type of error that could possibly be made. accept/reject decision and type of error next to each graph.

(2) Indicate and show on the graphs whether you would accept or reject the H_{0} for each of the three alpha values (.10, .05, .01) if **x-bar = 35.9** and the type of error that could possibly be made. accept/reject decision and type of error next to each graph.

### Recently Asked Questions

- What are the reasons why the 1998 Human Rights Act was really needed?

- how do I use the fzero function to solve this problem in matlab 2xy-40e^x=0 x^2-y=50 the solution for x lies near 0.

- Use the normal distribution of fish lengths for which the mean is 10 10 inches and the standard deviation is 2 2 inches. Assume the variable x is normally