DASC 120 Week 6 Lecture 6.2 One Sample Proportion Testing

Type 1 error - Rejecting the null when we should not have

Type II error - Failing to reject the null when we should have

Confidence Level - related to the alpha level

Alpha - the amount of risk we are wi ing to take of Type 1 error

p-obs - the calculated value from our data to compare to alpha

If the formal HT (Hypothesis Testing): limitations and conditions hold,

for single proportions

Remember sampling distributions use standard error!

Determining sample size for proportion

If p is unknown (most cases), we use phat in the calculation of the standard error

Choosing a sample size

If there isnt a previous study, use phat = 0.5

• If you don’t know any better, 50/50 is a good guess

• 0.5 gives the most conservative estimate -- highest possible sample size