Null Hypothesis

Null hypothesis

says no difference between means or no relationship among the variables

If the correlation coefficient is zero the slopes will be…

the same

If the means are equal to each other,

lines will be parallel

Accepting the null when true

specificity

Rejecting the null when false

sensitivity

Rejecting the null when true

Type 1 error

Accepting the null when false

Type 2 error

How do you avoid making a Type 1 error

set a low alpha level

T/F The p value needs to be lower than the alpha to be significant

True

T/F If there is no significance (p value > alpha), you should accept the null hypothesis

True