Statistics: Comparing 2 proportions

Question 1

What is meant by independent samples?

Answer 1

The observations in one sample are independent of those in another sample.

Question 2

When are two samples dependent?

Answer 2

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.

Question 3

What is a meta analysis?

Answer 3

Combines the results of several studies that address a set of statistical questions.

Question 4

When is the sampling distribution normal?

Answer 4

When it has a large sample

Question 5

What do negative signs mean in a confidence interval

Answer 5

P1 is less than P2 by that variable.

Question 6

What does it mean if the confidence interval contains 0?

Answer 6

There’s insufficient evidence to infer whether p1 or p2 is larger

Question 7

What is the format for a null hypothesis when comparing two groups?

Answer 7

H0: (p1- p2)=0

Question 8

What is meant by a pooled estimate

Answer 8

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.

Question 9

What does the test statistic measure?

Answer 9

the number of errors that the sample estimate falls from its null hypothesis

Question 10

What are the assumptions for a two sided confidence test for comparing two population proportions

Answer 10

Categorical for both groups, Independent random samples, five successes and failures in n1 and n2 if using two sided alternative

Question 11

What is the format for a one sided alternative hypothesis?

Answer 11

H0: (p1-p2) >/< 0

Question 12

If we’re comparing two proportions then the standard error of the difference is ________ than the standard error for either sample portion alone

Answer 12

Larger

Question 13

How do you estimate the df in the confidence interval between two population means

Answer 13

The smaller of (N1-1) and (N2-1)

Question 14

When can we not rely on the central limit theorem

Answer 14

When the sample size is under 30

Question 15

What assumptions are there in a confidence interval for the difference between two population means ?

Answer 15

Independent random samples and an approximately normal dist. in each group.

Question 16

List three possible interpretations of the confidence interval

Answer 16

If 0 falls in the interval it means M1=M2 is possible.
If it only contains positive numbers then it suggests that M1-M2 is positive therefore we infer that M1 is larger than M2
If it only contains negative numbers then it suggests that M1-M2 is negative therefore we infer that M1 is smaller than M2

Question 17

When are one sided t tests not trustworthy?

Answer 17

When the sample size is under 30 and the distribution is skewed.

Question 18

What is meant by the pooled standard deviation?

Answer 18

The combination of information from the two samples to provide a single estimate of variability.

Question 19

What is meant by the relative risk?

Answer 19

The ratio of proportions (p1/p2) in medical applications for which the proportions refers to an outcome which i undesirable, such as death or having a heart attack.

Question 20

What are the assumptions of a paired difference t test?

Answer 20

The sample of difference scores is a random sample from a population of difference scores and they have an approximately normal distribution.