Hypothesis Testing Flashcards
Null Hypothesis
H0, what we assume to be true
Alternative Hypothesis
H1, what would be true if H0 is wrong
Significance level
The given threshold of likeliness
One tailed test
When H1: p> k or P < k
Two tailed test.
When H1: p != k -> halve significance level for each enf.
Correlation hypothesis test
H0: p = 0
H1: p>0 or p<0 or p != 0
Compare your r with table
If your r is more extreme, reject H0
If your r is less extreme, no evidence to reject H0
REMEMBER CONTEXT
Binomial hypothesis test
Test statistic: number of successes observed.
H0: p = k
H1: p>k or p< k or P != l
Assume H0 i.e X varies B(n, k)
Find P(X >= or <= value in question) and more extreme
If P less than significance value, reject H0
If P greater than significance value, no evidence to reject H0
REMEMBER context