Epidemiology/Stats Flashcards

1
Q

For which study designs does one use risk ratio/rate ratio (relative risk)?

A

Calculated for study designs that collect data on incidence: cohort and RCT

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

For which study design does one use risk odds ratio?

A

Calculated for study designs that use prevalent cases: cross-sectional and case-control studies

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What is type I (alpha) error?

A

True null hypothesis incorrectly rejected (false positive)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is type II (beta) error?

A

Failing to reject false null hypothesis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Power equation

A

1 – beta

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

ROC axes?

A

Sensitivity on Y axis and 1-specificity on X axis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

ROC AUC?

A

AUC = area under curve (accuracy of test)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Left shift of ROC has what result?

A

Increased accuracy

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Parametric vs non-parametric tests in large samples

A

The nonparametric tests nearly as powerful as the parametric test from Gaussian populations

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Parametric vs non-parametric tests in small Gaussian samples

A

Non-parametric tests have much less power

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Figure out what this means…

A

 Usually report p values and not CIs because this requires additional assumptions
 Cannot be readily extended to regression models
 Consider transforming data to create a Gaussian distribution rather than use nonparametric and the decision of whether to use parametric vs nonparametric most important with small data sets

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Parametric and non-parametric tests for 2 continuous variables

A

T test and Mann-whitney U (Wilcoxon Rank Sum) respectively

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Parametric and non-parametric tests for 2 paired continuous variables

A

Paired t-test and Wilcoxon Signed Rank respectively

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Parametric and non-parametric tests for 3 continuous variables

A

ANOVA and Kruskal-Wallis respectively

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Parametric and non-parametric tests for 3 continuous variables repeat measurements

A

Repeated measures ANOVA and Friedman respectively

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Parametric and non-parametric tests for correlation of continuous variables

A

Pearson’s coefficient and Spearman’s coefficient respectively

17
Q

Tests for comparison

2 groups, independent

A

Chi-Squared, Fisher’s exact

18
Q

Tests for comparison

2 groups, dependent

A

McNemar’s Chi squared

19
Q

Tests for comparison

3 groups, independent

A

Chi-Squared, Fisher’s exact

20
Q

Tests for comparison

3 groups, dependent

A

Cochran’s Q test

21
Q

Association between two variables

A

Odds Ratio, Relative Risk

22
Q

Goal of regression analysis

A

To determine the values of parameters for a function that cause the function to best fit a set of data observations that you provide

23
Q

Variables/equation in linear regression

A

Both continuous variables, straight line equation

24
Q

Logistic regression

A

Probability of occurrence of an event by fitting data to a logistic curve; generalized linear model used for binomial regression. Makes use of several predictor variables that may be either numerical or categorical

25
Q

COX regression

A

Survival (multiple groups)