Choose
What is the null hypothesis for any 2-sample z test?
Ho: p1-p2 = 0
When you have random assignment, you do not need to check which condition?
P-value = .235; alpha = .05; reject or fail to reject the null?
p-value greater than alpha, so reject!
True or false: the power of a test is a probability
True!
Power = P (correctly reject Null | Alt is true)
= 1 - P(Type 2 error)
Mr. Otey claims that he is a 70% free throw shooter, but Kayla thinks he is not that good. State the null and alt hypotheses, and define the parameter.
Ho: P =.70
Ha: P<.70
P= true proportion of free throws that Mr. Otey makes
When checking the Large Counts condition for a one-sample z test, do you use p-hat or p-null?
P-null
Bonus: why?
We took a sample of skittles and a sample of M and M's and calculated the proportion of orange in each sample. Assume all conditions for inference were met. We ran a two-sided hypothesis test and we got a p-value of .07. At the .05 significance level, what should we conclude?
Since p-value is greater than alpha, we fail to reject the null and we do not have convincing evidence of a difference in proportions of orange between the two candies.
What is the probability of making a type 1 error?
Alpha!
Mr. Otey claims that he is a 70% free throw shooter, but Kayla thinks he is not that good. Mr. Otey shoots 40 free throws and makes 35 of them. How strong is the evidence in favor of the alternative hypothesis?
35/40 = 87.%, which is much greater than 70%. So there is NO evidence in favor of the alternative hypothesis! No need to even run a hypothesis test.
You took a random sample of 30 students at a school and asked them whether they have a driver's license. In order to run a one-sample z test, what needs to be true about the student population?
N >= 300 (10% condition)
Bonus: suppose your null hypothesis is that p0 =.3. Can you run your hypothesis test?
We took a sample of skittles and a sample of M and M's and calculated the proportion of orange in each sample. Assume all conditions for inference were met. We ran a two-sided hypothesis test and we got a p-value of .07. At the .10 significance level, what should we conclude?
Since p-value is less than alpha, we reject the null and we do have convincing evidence of a difference in proportions of orange between the two candies. (state the direction of the difference)
The three ways to increase the power of a test are...
1) Larger sample
2) Larger Alpha
3) Greater distance between the actual parameter value and the hypothesized parameter value
If we want to know whether LCA or BTA has a higher proportion of seniors taking AP classes, what test should we use?
2-sample z test (for difference of proportions)
For a two-sample z test, when checking the large counts condition, do use p-hat or p-null?
Trick question! You use p-pooled
The manager of an ice cream shop claims that 80 percent of his customers are very satisfied with the distribution of chunks in the ice cream. A local newspaper thinks he is exaggerating, so they survey 100 randomly selected customers. 73 of those surveryed say they are very satisfied. Based on these findings, give the test statistic and p-value.
z = -1.75, p-value = .04
Which of the three p's can be a negative number?
In disease testing, a false negative corresponds to what kind of error?
Type 2 error!