Logic and Vocabulary of Hypothesis Tests Chapter 13
Hypothesis Testing Logic Assume a value for a parameter. Collect sample data to test the assumption. Draw a conclusion about the parameter.
Vocabulary The Null Hypothesis (H o ): The statement about a parameter that we assume is true. It must contain equality. It is the statement that is tested.
Vocabulary The null hypothesis is the statement of ‘no effect’ or ‘no difference’ from the null value.
Vocabulary The Alternative Hypothesis (H a ): The “research” statement. It is usually what we think might be true and hope to find evidence to support.
Vocabulary If we find evidence against H o, then we are in favor of H a. If H o is rejected, then the test is ‘statistically significant’.
Vocabulary The Test Statistic is a standard score, computed from data and based on assuming H o is true.
Vocabulary The P-value is the probability of observing our statistic or one more extreme in the direction of H a. The smaller the P-value, the stronger the evidence against H o.
Vocabulary The P-value is the area in the appropriate tail(s) of the sampling distribution. The tail desired depends on H a.
Vocabulary The significance level is denoted by . We will reject H o when the P-value is less than . We’ll usually use = 0.05 but other common values are 0.01 or 0.10.
Hypothesis Testing Procedure Define parameters and state H o and H a. Check conditions. Calculate the test statistic. Find the P-value. Write a conclusion in context.