1. SECTION 1 TEST OF A SINGLE PROPORTION
If the data we are analyzing are nominal data, the hypothesis might be a statement about:
The value of the proportion, , of population members that have a certain characteristic (one of the categories of the nominal variable).

2. SECTION 1 TEST OF A SINGLE PROPORTION
For example, the hypothesis might be a statement about the proportion of
Students who are interested in graduate school
Vaccinated patients who remain cancer free
CEOs who use computers as a major tool
People who are unemployed

3. SECTION 1 TEST OF A SINGLE PROPORTION
Two-Tail Tests of Proportions
We will use precisely the same five steps that we have been following for any hypothesis test:
Step 1: Set up the null and alternative hypotheses test
Step 2: Pick the value of  and find the rejection region.
Step 3: Calculate the test statistic.
Step 4: Decide whether or not to reject the null hypothesis.
Step 5: Interpret the statistical decision in terms of the stated problem.

4. SECTION 1 TEST OF A SINGLE PROPORTION
Five-step hypothesis testing procedure is identical to the one we have been using.
The test statistic is the same as that used for a two-tail test of proportions and the rejection regions are the same as those used for one-tail tests of the mean.

5. SECTION 1 TEST OF A SINGLE PROPORTION
One-Tail Test of Proportions
Step 1: Set up the null and alternative hypotheses.
There are two possible ways to set up a one-tail test of proportions.
Upper-Tail Test
Ho:   [a specific number]
HA:  > [a specific number]

6. SECTION 1 TEST OF A SINGLE PROPORTION
Lower-Tail Test
Ho:  [a specific number]
HA: < [a specific number]
Step 2: Select the value of “” and find the rejection region.
Step 3: Calculate the test statistic.

7. SECTION 1 TEST OF A SINGLE PROPORTION
We have seen in the previous section that the appropriate test statistic is
Steps 4 and 5 remain the same.

11. SECTION 1 COLLECTING DATA FROM TWO POPULATIONS
P-level
The value of the p-level represents a decreasing index of the reliability of a result.
The higher the p-level, the less we can believe that the observed relation between variables in the sample is a reliable indicator of the relation between the respective variables in the population.

12. SECTION 1 COLLECTING DATA FROM TWO POPULATIONS
Specifically, the p-level represents the probability of error that is involved in accepting our observed result as valid, that is, as "representative of the population."
For example, the p-level of .05 (i.e.,1/20) indicates that there is a 5% probability that the relation between the variables found in our sample is a "fluke."

13. SECTION 1 COLLECTING DATA FROM TWO POPULATIONS
In many areas of research, the p-level of .05 is customarily treated as a "border-line acceptable" error level.
For more info:

14. Problem 17 (Page 65 or 153)
Fowle Marketing Research Inc., bases charges to a client on the assumption that telephone surveys can be completed in a mean time of 15 minutes or less. If a longer mean survey time is necessary, a premium rate is charged. Suppose a sample of 35 surveys shows a sample mean of 17 minutes and a sample standard deviation of 4 minutes. Is the premium rate justified?

15. Problem 17 (Page 65 or 153)
Formulate the null and alternative hypotheses for this application.
Compute the value of the test statistic.

16. Using =0.01, what is your conclusion?
Problem 17 (Page 65 or 153)
What is the p -value?
Using =0.01, what is your conclusion?
Since P-value= < = 0.01, we reject H0, that means the premium rate is justified.

17. SECTION 1 COLLECTING DATA FROM TWO POPULATIONS
When you are comparing characteristics of two different populations, you must have a sample from each of the populations. These samples are usually selected independently of each other.
In other words, the selection of one sample should not have any effect on the selection of the second sample.

18. SECTION 1 COLLECTING DATA FROM TWO POPULATIONS
We will label all of the parameters of one population with a subscript 1 and all the parameters of the second population with a subscript 2.
It does not matter which population you label 1 or 2. The populations and samples are shown in Figure 13.1.

20. SECTION 1 COLLECTING DATA FROM TWO POPULATIONS
Consider the question posed about whether men or women spend more money on frozen foods.
We could label the population of males as population 1 and the population of females as population 2.

21. SECTION 1 COLLECTING DATA FROM TWO POPULATIONS
If we do this then the parameters and statistics corresponding to the male population will be identified with a subscript 1 and those describing the female population will carry a subscript 2.
For this example, we would select a sample of men shoppers and a separate sample of women shoppers.

22. SECTION 1 COLLECTING DATA FROM TWO POPULATIONS
We would ask all members of the sample how much money they spent on frozen foods in the past week. It is not necessary that the sample sizes be equal, but if possible it is desirable to have both sample sizes (n1, n2) greater than or equal to 30.
The reason for this stems from the fact that the Central Limit Theorem generally applies when the sample size is 30 or greater.

23. SECTION 1 COLLECTING DATA FROM TWO POPULATIONS
Remember that we developed the Z test statistic based on the knowledge that the sample mean, (X-bar), has a normal distribution.
However, often a single sample is selected and a qualitative variable is used to identify two populations for comparison. For the food shopper example, we might select one sample of shoppers and then record the gender of the respondent as part of the data.

24. SECTION 1 COLLECTING DATA FROM TWO POPULATIONS
This means that the data can then be divided into the two comparison-populations after the data have been collected.
If at the same time you collect data on the age of the person as "under 40" or "40 and over" then you can also compare the average frozen food expenditure for younger buyers to that of older buyers.

25. SECTION 1 COLLECTING DATA FROM TWO POPULATIONS
Clearly, spending differences that are identified by gender or age could be of great assistance in developing a marketing strategy.

26. SECTION 1 HYPOTHESIS TEST OF THE DIFFERENCE IN TWO POPULATION MEANS
Regardless of the particular case, each hypothesis test will follow the five-step procedure
Step 1: Set up the null and alternative hypotheses.
Step 2: Pick the value of  and find the rejection region.
Step 3: Calculate the test statistic and the p value.
Step 4: Decide whether or not to reject the null hypothesis.
Step 5: Interpret the statistical decision in terms of the stated problem.

27. SECTION 1 LARGE-SAMPLE TESTS OF THE DIFFERENCE IN TWO POPULATION MEANS
Large-Sample Tests of Two Means with Known
Standard Deviations
The basic test concerning two population means occurs when we want to know whether the two samples come from populations with equal means and we assume that the population standard deviations are known.

28. SECTION 1 LARGE-SAMPLE TESTS OF THE DIFFERENCE IN TWO POPULATION MEANS
The first step of the procedure is to construct the null and alternative hypotheses. As with tests of a single mean, there are three different ways to set up the hypothesis test.
Notice that the equals sign is always part of the null hypothesis and hypotheses are statements about the relationship between the size of the mean of population 1 and the mean of population 2.

29. SECTION 1 LARGE-SAMPLE TESTS OF THE DIFFERENCE IN TWO POPULATION MEANS
The tests do not give you information about the value of the means, only about how the value of 1 compares to the value of 2.
You can see that there are two ways to state each of the hypotheses.
For each setup shown above the first way of writing the test makes a statement about the relative value of 1 to 2.

30. SECTION 1 LARGE-SAMPLE TESTS OF THE DIFFERENCE IN TWO POPULATION MEANS
The second way of writing the same test makes a statement about the value of the difference, 1-2.
They are equivalent to each other. Look at the two-sided test.
Clearly, if 1=2 then the difference between them must be zero. It is also possible to test for differences of values other than zero.

31. SECTION 1 LARGE-SAMPLE TESTS OF THE DIFFERENCE IN TWO POPULATION MEANS
Large-Sample Tests of Two Means with Unknown
Standard Deviations
The large-sample test for the difference between two population means requires that both of the sample sizes be greater than 30.
Since the sample sizes are large, each individual sample standard deviation is a good estimate of the corresponding unknown population standard deviation.

32. SECTION 1 LARGE-SAMPLE TESTS OF THE DIFFERENCE IN TWO POPULATION MEANS
So, we simply use each of the sample standard deviations in the formula instead of the corresponding values of σ.
The test statistic becomes
The rejection region depends on  and whether the test is one-sided or two-sided.