Chapter 6 Flashcards
Five Steps of Hypothesis/Significance Test
- Making Assumptions
- Constructing Hypothesis
- Determining Test Statistic
- Calculating the p-value
- Drawing Conclusion
Making Assumptions
Assume the variable of interest (binary, quantitative?) sample type (simple random), distribution (normal)
Constructing Hypothesis
Two hypotheses: Null and Alternate
Null Hypothesis
H0, statement specifying particular parameter value for either pop. mean or proportion, determined from prior belief as a baseline
Alternate Hypothesis
Ha, alternative range of values for population parameter, either one or two-sided
One-Sided
Ha >, <
Two-Sided
Ha =/=
Test Statistic
Measures how close point estimate of parameter is to null hypothesis value, in terms of standard error
Test Statistic Formula
(Parameter Estimate - Null Hypothesis Value) / Standard Error
P-Value
Probability of getting a test statistic value more extreme than what we have actually got given that the null hypothesis is true; how plausible is the null hypothesis?
Small p-value
Test statistic takes value unlikely under null hypothesis; evidence against, for rejecting H0
Large p-value
Evidence for not rejecting H0
Drawing Conclusion
Decision of whether or not to reject H0 by comparing p-value and significance level alpha
If p-value less than or equal to alpha
We reject H0
If p-value greater than alpha
We fail to reject H0
