T-tests

Question 1

What comparison will you use when evaluating continuous data with a normal distribution?

Answer 1

The means. Use the unpaired or paired t-test.

Question 2

What comparison will you use when evaluating continuous data without a normal distribution?

Answer 2

The medians. Mann-Whitney WIlcoxon test (unpaired) or signed rank test (paired)

Question 3

What comparison will you use when evaluating categorical data with ordinal data?

Answer 3

Medians or proportions. Use medians for larger sample size.

Question 4

What comparison will you use when evaluating categorical data that is not ordinal?

Answer 4

Proportions. Z test/Chi square.

Question 5

What is a classic case of paired data?

Answer 5

Measuring the same person twice (pre and post measurements)

Question 6

Cross-over trial. THIS WILL BE ON TEST.

Answer 6

Patients get placebo treatment and drug treatment and reactions are compared. Order is randomized.

Question 7

Matched case-control study

Answer 7

Measuring two people from similar demographics

Question 8

Twin study

Answer 8

Measuring reactions between identical twins

Question 9

Unpaired t-test

Answer 9

Question 10

Mann-Whitney Wilcoxon Test

Answer 10

Used to compare medians for data without a standard distribution.

Question 11

Nonparametric test

Answer 11

Tests where you are not worried about the distribution. Good with ordinal data. You don’t have to worry about outliers. Less powerful than a t-test in normal distributed data.

Question 12

Paired t-test

Answer 12

Question 13

Wilcoxon signed rank test

Answer 13

Used with paired continuous data that is not normally distributed

Question 14

Z test

Answer 14

difference in proportions/pooled standard error

Question 15

Chi square test

Answer 15

((observed frequencies-expected frequencies)^2)/ expected

Question 16

How can you get the Z statistic from the Chi Square statistic?

Answer 16

Take the square root of your Chi Square value

Question 17

Why do you use a Chi Square test?

Answer 17

Comparing unpaired proportions. To prove that two proportions are the same. Use in RR, OR, AR and prevalence ratio.

Question 18

Fisher’s exact test

Answer 18

Comparing unpaired proportions. Use for 2x2 tables dealing with small samples. Asks how many 2x2 table values are likely to be more extreme than the values you got.

Question 19

Paired and unpaired tests

Answer 19

Question 20

McNemar’s Chi Square

Answer 20

Determines the ratio of people switching from no symptoms to symptoms. Allows for prevalence comparison.

Question 21

Comparing unpaired data in 3+ groups

Answer 21

Use analysis of variance for means, Kruskal-Wallis for medians and Chi square for proportions

Question 22

Multiple comparisons problem

Answer 22

The more t-tests you do with an alpha of .05 accumulates the probability of making a type I error.

Question 23

Bonderroni adjustment

Answer 23

Says you are okay with 5% probability of error across the entire test. So in a study of 12 different tests your criteria becomes more strict (5%/12). Increases probability of type II error.

Question 24

p-values

Answer 24

Says you are unlikely to get results due to chance. You could still get your result due to bias. Based on clinical significance, not clinical importance.