T-tests Flashcards Preview

Fundamentals > T-tests > Flashcards

Flashcards in T-tests Deck (24)
Loading flashcards...
1
Q

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

A

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

2
Q

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

A

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

3
Q

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

A

Medians or proportions. Use medians for larger sample size.

4
Q

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

A

Proportions. Z test/Chi square.

5
Q

What is a classic case of paired data?

A

Measuring the same person twice (pre and post measurements)

6
Q

Cross-over trial. THIS WILL BE ON TEST.

A

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

7
Q

Matched case-control study

A

Measuring two people from similar demographics

8
Q

Twin study

A

Measuring reactions between identical twins

9
Q

Unpaired t-test

A
10
Q

Mann-Whitney Wilcoxon Test

A

Used to compare medians for data without a standard distribution.

11
Q

Nonparametric test

A

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.

12
Q

Paired t-test

A
13
Q

Wilcoxon signed rank test

A

Used with paired continuous data that is not normally distributed

14
Q

Z test

A

difference in proportions/pooled standard error

15
Q

Chi square test

A

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

16
Q

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

A

Take the square root of your Chi Square value

17
Q

Why do you use a Chi Square test?

A

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

18
Q

Fisher’s exact test

A

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.

19
Q

Paired and unpaired tests

A
20
Q

McNemar’s Chi Square

A

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

21
Q

Comparing unpaired data in 3+ groups

A

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

22
Q

Multiple comparisons problem

A

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

23
Q

Bonderroni adjustment

A

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.

24
Q

p-values

A

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.