Hypothesis testing is used to determine whether a statement about a population parameter should be rejected.
Null Hypothesis (H0): A tentative assumption about a population parameter.
Example Question: A company claims that its batteries last for 500 hours. What would be the null hypothesis to test this claim?
Alternative Hypothesis (Ha): The opposite of the null hypothesis.
Example Question: If the null hypothesis is that the average height of women is 5'4", what could be an alternative hypothesis?
The hypothesis testing procedure uses sample data to test the two competing statements indicated by H0 and Ha.
Developing Null and Alternative Hypotheses
Formulating hypotheses requires careful structuring to ensure the test conclusion provides the desired information.
The context of the situation is very important in determining how the hypotheses should be stated.
Sometimes it's easier to identify the alternative hypothesis first, while other times the null hypothesis is easier to define initially.
Correct hypothesis formulation requires practice.
Alternative Hypothesis as a Research Hypothesis
Many hypothesis testing applications involve gathering evidence to support a research hypothesis.
It is often best to begin with the alternative hypothesis, making it the conclusion the researcher hopes to support.
The research hypothesis is considered true if the sample data provides sufficient evidence to reject the null hypothesis.
Example 1: A new teaching method is believed to be better than the current one.
Alternative Hypothesis: The new teaching method is better.
Null Hypothesis: The new method is no better than the old method.
Example Question: If a new teaching method is implemented, what hypothesis would a researcher aim to support?
Example 2: A new sales force bonus plan is developed to increase sales.
Alternative Hypothesis: The new bonus plan increases sales.
Null Hypothesis: The new bonus plan does not increase sales.
Example Question: How would you formulate the alternative hypothesis to test if a bonus plan increases sales?
Example 3: A new drug aims to lower blood pressure more than the existing drug.
Alternative Hypothesis: The new drug lowers blood pressure more than the existing drug.
Null Hypothesis: The new drug does not lower blood pressure more than the existing drug.
Example Question: What null hypothesis would you use to test if a new drug lowers blood pressure more effectively?
Null Hypothesis as an Assumption to be Challenged
Start with a belief or assumption about a population parameter's value.
Use a hypothesis test to challenge the assumption and determine if there's statistical evidence to conclude the assumption is incorrect.
In these situations, it's helpful to develop the null hypothesis first.
Example: The label on a soft drink bottle states that it contains 67.6 fluid ounces.
Null Hypothesis: The label is correct (μ=67.6 ounces).
Alternative Hypothesis: The label is incorrect (μ=67.6 ounces).
Example Question: If you want to challenge the accuracy of a soft drink bottle's label, what initial hypothesis would you set?
Forms for Null and Alternative Hypotheses about a Population Mean
The equality part of the hypotheses always appears in the null hypothesis.
A hypothesis test about a population mean μ must take one of the following three forms (where μ0 is the hypothesized value of the population mean):
One-tailed (lower-tail):
H<em>0:μ≥μ</em>0
H_a: \mu - One-tailed (upper-tail):
H0: \mu \le \mu0
Ha: \mu > \mu0
Example Question: Give an example of when you will use one tailed upper test.
Two-tailed:
H0: \mu = \mu0
Ha: \mu \neq \mu0
Example Question: Give an example of when you will use two tailed test.
Example: Metro EMS
A major west coast city's emergency medical service aims to respond to medical emergencies with a mean time of 12 minutes or less.
The director wants to formulate a hypothesis test to determine whether the service goal of 12 minutes or less is being achieved, using a sample of emergency response times.
Null and Alternative Hypotheses:
H_0: \mu \le 12 (The emergency service is meeting the response goal; no follow-up action is necessary.)
H_a: \mu > 12 (The emergency service is not meeting the response goal; appropriate follow-up action is necessary.)
Where \mu = mean response time for the population of medical emergency requests.
Example Question: If the EMS wants to ensure they are meeting their response time goals, how should they set up their null and alternative hypotheses?
Type I and Type II Errors
Hypothesis tests are based on sample data, so errors are possible.
Type I Error: Rejecting H0 when it is true.
The probability of making a Type I error when the null hypothesis is true as an equality is called the level of significance, denoted by \alpha.
Hypothesis testing that only controls the Type I error are called significance tests.
Example Question: What is the consequence of rejecting a true null hypothesis?
Type II Error: Accepting H0 when it is false.
It is difficult to control for the probability of making a Type II error.
Statisticians avoid the risk of making a Type II error by using "do not reject H0" and not "accept H0".
Example Question: Why do statisticians prefer to say "do not reject H0" instead of "accept H0"?