Stats

Question 1

formula for NNT

Answer 1

NNT = 1/ARR
ARR = absolute risk reduction

Question 2

type I error

Answer 2

rejecting the null hypothesis when it’s actually correct
false positive study
alpha

Question 3

type II error

Answer 3

failure to reject the null hypothesis appropriately
false negative study
beta

Question 4

failure to reject the null hypothesis appropriately

Answer 4

type II error

Question 5

rejecting the null hypothesis when it’s actually correct

Answer 5

type I error

Question 6

What 3 factors do you need to look at to determine the appropriate type of statistical test to perform?

Answer 6

1. is the distribution parametric or not?
2. is your data continuous, nominal/ordinal, or categorical?
3. are your tested populations dependent or independent?

Question 7

Sensitivity =

Answer 7

TP / TP+FN

Question 8

Specificity =

Answer 8

TN / TN+FP

Question 9

Power

Answer 9

1 - beta

beta = type II error rate

Question 10

PPV =

Answer 10

TP / TP+FP

Question 11

NPV =

Answer 11

TN / TN+FN

Question 12

Incidence vs Prevalence

Answer 12

Incidence = RATE of new diseases over a period of time
Prevalence = number of existing disease cases at a specific POINT in time

Question 13

What do you need to calculate the sample size for a study?

Answer 13

1. the effect size
2. the type I error rate
3. the type II error rate

Question 14

what is the reciprocal of the rate difference?

Answer 14

the NNT

Question 15

What is the other name for the Mann-Whitney U test?

Answer 15

Wilcoxon rank-sum test

Question 16

What is another name for the Wilcoxon rank-sum test?

Answer 16

Mann-Whitney U test

Question 17

What is the t-test?

Answer 17

statistical test for parametric, continuous data that is independent

Question 18

what is the paired t-test?

Answer 18

statistical test for parametric, continuous data that is paired

Question 19

what is the ANOVA?

Answer 19

statistical test for parametric, continuous data with 3 or more independent groups

Question 20

what is the wilcoxon signed-rank sum test?

Answer 20

statistical test for non-parametric, continuous data that is paired
non-parametric corollary to paired t-test

Question 21

what is the wilcoxon ranked-sum test?

Answer 21

statistical test for non-parametric, continuous data that is independent
nonparametric corollary to t-test
also called the mann-whitney u test

Question 22

what is the kruskal-wallis?

Answer 22

statistical test for non-parametric, continuous data with 3 or more independent groups
nonparametric corollary to ANOVA

Question 23

what is the Chi-square test?

Answer 23

statistical test for parametric, categorical data with independent groups

Question 24

what is the Fisher’s exact test?

Answer 24

statistical test for parametric, categorical data with independent groups if there are <5 measurements/group

Question 25

what is logistic regression?

Answer 25

statistical test to predict the relationship between a DICHOTOMOUS outcome vs a set of variables while controlling for other variables in the analysis

Question 26

what is Bonferroni's correction

Answer 26

a method to correct for multiple repeated testing on the same data set

Question 27

what is Kolmogorov-Smirnov?

Answer 27

test to determine if data is parametric or not

Question 28

what is Shaprio-Wilk?

Answer 28

test to determine if the data is parametric or not

Question 29

LR +

Answer 29

= (positive test/presence of disease) / (positive test/absence of disease) = sensitivity / (1 - specificity)

Question 30

LR -

Answer 30

= (negative test/presence of disease) / (negative test/absence of disease) = (1 - sensitivity) / specificity

Question 31

How do you use LRs?

Answer 31

You have to convert the pretest probability to odds This is pretest probability / (1 - pretest probability) Then you multiply by the LR Finally, you convert the odds BACK to the probability!

Question 32

What are clinically significant LRs?

Answer 32

LR + > 10 | LR - < 0.1

Question 33

formula for odds ratio

Answer 33

(AxD)/(BxC) in standard 2x2 table

Question 34

what's the difference between odds ratio and relative risk?

Answer 34

relative risk is used when patients are followed over time. | odds ratio is used when patients already have the outcome and you look back retrospectively at an exposure of interest