Statistics: Comparing 2 proportions Flashcards
What is meant by independent samples?
The observations in one sample are independent of those in another sample.
When are two samples dependent?
When they have the same subjects, if they’re in matched pairs (each subject in one sample is matched with a subject in the other sample,) or when each sample is observed twice.
What is a meta analysis?
Combines the results of several studies that address a set of statistical questions.
When is the sampling distribution normal?
When it has a large sample
What do negative signs mean in a confidence interval
P1 is less than P2 by that variable.
What does it mean if the confidence interval contains 0?
There’s insufficient evidence to infer whether p1 or p2 is larger
What is the format for a null hypothesis when comparing two groups?
H0: (p1- p2)=0
What is meant by a pooled estimate
one that pools the total number of successes and total number of observations from the two samples in the format s1+s2/n1+n2 where s= successes.
What does the test statistic measure?
the number of errors that the sample estimate falls from its null hypothesis
What are the assumptions for a two sided confidence test for comparing two population proportions
Categorical for both groups, Independent random samples, five successes and failures in n1 and n2 if using two sided alternative
What is the format for a one sided alternative hypothesis?
H0: (p1-p2) >/< 0
If we’re comparing two proportions then the standard error of the difference is ________ than the standard error for either sample portion alone
Larger
How do you estimate the df in the confidence interval between two population means
The smaller of (N1-1) and (N2-1)
When can we not rely on the central limit theorem
When the sample size is under 30
What assumptions are there in a confidence interval for the difference between two population means ?
Independent random samples and an approximately normal dist. in each group.