Confidence Intervals for Proportions
Estimate +- "Margin of Error"
What is the formula for a confidence interval?
The starting hypothesis to be tested
What is the null hypothesis H_0 ?
The error of rejecting a null hypothesis when it is true
What is a Type I error, or a false positive, or a false alarm?
P(observed statistic value [or more extreme] | H_0 is true)
What is the P-value?
The 2 different independence assumptions that need to be true to compare 2 proportions
What are the Independent Groups Assumption and the Independence Assumption?
sqrt((hat p hat q)/n)
What is the standard error for the sampling distribution?
It contains all the values of the parameter that we'd consider plausible if we reject the null hypothesis
What is the alternative hypothesis H_A ?
The error of not rejecting a null hypothesis when it is false
What is a Type II error, or a false negative?
Formula to calculate the power of a test
What is 1 - beta?
sqrt((hat p_1 hat q_1)/n_1+(hat p_2 hat q_2)/n_2)
What is SE(hatp_1-hatp_2)?
The number of standard errors to move away from the sample statistic to specify an interval that corresponds to the specified level of confidence
What is the critical value?
The probability of seeing data like these (or something even less likely) given that the null hypothesis is true
What is the P-value?
The probability of rejecting a null hypothesis when it is true
What is the significance level, or alpha ?
A pro and a con of increasing a test's significance level alpha
What are the pros: test has more power, fewer Type II errors
What are the cons: more Type 1 errors
("Success"_1 + "Success"_2)/(n_1 + n_2)
What is the formula for hatp_"pooled" ?
z * sqrt((hat p hat q)/n)
What is the formula for the margin of error?
A way to proceed when there is no obvious sampling distribution model to use, and there may be no data available
What is a simulation?
The probability of not rejecting a null hypothesis when it is false
What is beta?
Two ways (out of 4 given in the text) of increasing a test's power
Teacher discretion here
hat p_a-hat p_b +- z * SE(hat p_a-hat p_b)
What is the formula for a confidence interval for the difference between two proportions?
hat p +- z * SE(hat p)
What is the formula for a confidence interval?
The difference between the value you see in your data and the value in the null hypothesis
What is the effect size?
The probability that a test will correctly reject a false null hypothesis
What is the power of the test?
Another way (not including the two ways mentioned previously, out of 4 given in the text) of increasing a test's power
Teacher discretion here
z=((hat p_1 - hat p_2) - 0)/("SE"_"pooled"(hat p_1- hatp_2))
What is the test statistic z that you use to do a hypothesis test on the difference between 2 proportions?