Hypothesis Testing Basics Flashcards
Step 1
State the null hypothesis
Step 2
State the alternative (experimental) hypothesis
Step 3
Set the alpha (significance) level (α = .05 (or α = .01)
Step 4
Set the rejection rule
Step 5
Compute your statistic
Step 6
Decision
Alpha Level
Your alpha level is your willingness to make a type I error/results due to chance (traditionally 1% or 5%).
Ho
The results from our study are solely to chance.
Ha (H1)
The results from our study are due to something else (relationship/effect of I.V.)
Decision Rule
Whatever we set alpha level to (.05/.01), there is a cutoff to decide whether or not there is a relationship.
Directional (one-sided
When you specify a particular direction that you expect your results to fall in.
Nondirectional (two-sided)
Not predicting a certain direction, just indicating that there will be a difference.
Type I Error
Accepting the alternative hypothesis when there is no relationship (innocent found guilty → false alarm)
Type II Error
Accepting Ho when there is a relationship (guilty person found innocent → miss)