Statistical Hypothesis Testing Flashcards
What is the null hypothesis?
The null hypothesis is that ‘p’ takes a particular value which you would normally expect, or the original value
What is the alternative hypothesis?
The alternative hypothesis is that p > r or that p < r, where ‘r’ is what you would expect, or the original value
What is the test statistic?
The test statistic is a value found from experimental data
What is a hypothesis test?
A hypothesis test is a process in which you use statistical techniques to test a particular claim or theory
What is the p-value?
The p-value is the probability that you obtain a value at least as extreme as the test statistic
What is the significance level?
The level of probability used to reject the null hypothesis or not
What can be concluded if the p-value is less than the significance level?
Reject the null hypothesis
To get this value from the test statistic is so unlikely p-value of null hypothesis must be wrong
What can be concluded if the p-value is greater than the significance level?
Do not reject null hypothesis
Test statistic is not unlikely enough to suggest p-value of null hypothesis is wrong
How to remember whether to reject null or not?
If p is low, reject the H0
What is important to remember about the conclusion?
Non-assertive
Refers to the evidence
Always in terms of null hypothesis
What are the steps to a hypothesis test?
Determine the tail
Define test statistic and parameters
Define null and alternate
Calculate test statistic
Compare test statistic to significance level
Support or Reject
State conclusion using parameter
What is the difference between a one and two tail test?
Half the significance level for two tail test
What is the critical region?
The critical region is the range of values of the test statistic that would lead you to reject the null
How do you calculate the critical values?
Use the significance level to determine the percentage at which the critical region lies
Substitute ‘X’ values till it exceeds the critical region
What is the critical value?
The values on the boundary of the critical region
How do you calculate the actual significance level?
After determining where the critical region is, calculate the probability of X lying in that region
This is the actual significance level