Epi 3: Selecting Statistical Tests

Question 1

List 4 different categories that answer question # 2 in selecting correct statistical test.

Q2 = What type of comparison/assessment is desired?

Answer 1

• Correlation - 2 groups
• Regression - prediction, 3+ groups
• Survival Comparison - time
• Group Comparison
  2 vs. 3+ groups;
  independent vs. paired/related

Question 2

“prediction” indicates which category of statistical tests

Answer 2

Regression

3 regression tests = Logistic Regression, Multinominal Logistic Regression, Linear Regression

Question 3

What does correlation statistical test compare?

Answer 3

Quantitative representation of STRENGTH + DIRECTION of a relationship between TWO variables

FIXED value range = [-1 - +1]

\+/- = direction 
value = magnitude of strength

Question 4

How many Correlation statistical tests? Are any specific to particular data level?

Answer 4

Contingency Coefficient (nominal)
Spearman Correlation (ordinal)
Pearson Correlation (interval)
Kappa Statistic

Question 5

For Pearson Correlation test, what does a statistically significant value (p < 0.05) and a non-statistically significant value (p > 0.05) indicate?

Answer 5

p < 0.05: Linear correlation is present between 2 groups

p > 0.05: NO linear correlation, but may be a non-linear relationship present!!!

Question 6

Under what circumstances would you run a Pearson Correlation as a “partial correlation”? What level of data would this involve?

Answer 6

Partial Correlation = correlation stat. test that controls for confounding variables

Only interval data = Pearson Correlation

Question 7

Contingency Coefficient

Answer 7

Correlation stat. test

Nominal data

Question 8

Spearman Correlation

Answer 8

Correlation statistical test

Ordinal data

Question 9

Pearson Correlation

Answer 9

Correlation stat. test

Interval data

Question 10

Kappa Statistic

Answer 10

A correlation; measures STRENGTH + DIRECTION of agreement between different evaluators;

Compares ability of different evaluators to classify/interpret same data into 1 of many groups.
Ex.- 2 different radiologists interpret/classify the same radiographic image

FIXED value range: [-1 - + 1]

+/- = good agreement/poor agreement

Question 11

What does a regression measure? What else can be calculated from a regression?

Answer 11

Measure of relationship between multiple variables (IV) specific ability to predict outcome variable (DV)

Odds ratio for measure of association + control for confounding variables

Question 12

How determine the data level of a Regression?

Answer 12

= data level of outcome variable (DV) = variable being predicted

Question 13

What type of data comparison group can calculate Odds Ratio (OR)?

Answer 13

regression

Question 14

Nominal data + regression test =

Answer 14

Logistic Regression

Question 15

Ordinal data + Regression test =

Answer 15

Multinominal Logistic Regression

Question 16

Interval data + Regression test =

Answer 16

Linear Regression

Question 17

Logistic Regression

Answer 17

multiple independent variables used to predict outcome/dependent variable (DV)

predicted outcome data level = nominal

Question 18

Multinominal Logistic Regression

Answer 18

multiple independent variables used to predict outcome/dependent variable (DV)

predicted outcome data level = ordinal

Question 19

Linear Regression

Answer 19

multiple independent variables used to predict outcome/dependent variable (DV)

predicted outcome variable data level = interval

Question 20

List the regression test for each data level.

Nominal, ordinal, and interval data

Answer 20

logistic regression
multinominal logistic regression
linear regression

Question 21

For Pearson Correlation test, what does a non-significant value indicate?
(p > 0.05)

Answer 21

No LINEAR correlation between groups; however, may be NON-LINEAR correlations present

Question 22

delineate correlation vs. regression types of statistical tests

Answer 22

Correlation: relationship between TWO groups; fixed value range

Regression: relationship MULTIPLE IV in predicting an outcome variable (DV);

Question 23

what type of statistical test will only produce values within [-1 - +1]?

Answer 23

Correlation

Question 24

Partial Correlation

Answer 24

Correlation test ONLY for interval data (Pearson Correlation) that controls for confounding variables

Question 25

List all statistical tests that can be represented by Kaplan-Meier Curve

Answer 25

Log-Rank test Cox-Proportional Hazards test Kaplan-Meier test All survival tests

Question 26

"over time" indicates what type statistical comparison

Answer 26

survival test

Question 27

Survival test =

Answer 27

Compares proportion of events over time, or time-to events, between groups ~ongoing progression~

Question 28

"free from an event" over time

Answer 28

survival

Question 29

Log-Rank test

Answer 29

survival test + nominal level data | died/didn't die or developed disease/didn't develop disease

Question 30

Cox-Proportional test

Answer 30

survival test + ORDINAL level data collected ex. - tumor growth - 0 in., 1-5 in., 5-10 in., etc.

Question 31

Kaplan-Meier TEST

Answer 31

survival test + INTERVAL level data

Question 32

List all survival statistical tests for each data level:

Answer 32

``` nominal = Log-rank ordinal = cox-proportional hazards test interval = kaplan -meier test ```

Question 33

What is the most commonly utilized type of comparison of statistical data?

Answer 33

Group comparisons

Question 34

Which Correlation test can control for confounding factors?

Answer 34

Partial Correlation of Pearson Correlation test

Question 35

What kinds of comparisons are made in group comparisons?

Answer 35

frequencies counts proportions

Question 36

Type of statistical test assess event-occurrence and time-to-event?

Answer 36

survival tests

Question 37

What data being compared in Paired/Related Group Comparisons?

Answer 37

Comparing data of subjects from the SAME study groups

Question 38

"pre- vs. post-" "before vs. after" "baseline vs. end" should all indicate what kind of data?

Answer 38

Paired or related data - group comparisons

Question 39

What are the required assumptions of interval/ratio data for proper selection of parametric test?

Answer 39

- Normally-distributed - Equal Variances - Randomly-derived + Independent

Question 40

How do you handle interval/ratio data NOT normally-distributed?

Answer 40

1) TRANSFORM data --> Standardized Values (z-score/Log transformation) which might be Normally-distributed 2) Use Non-Parametric tests --> use equivalent stat test of the lower data level * Always run descriptive statistics and graphs*

Question 41

Null Hypothesis

Answer 41

Research perspective that states there will be NO true differences between groups being compared **p value** of statistical tests is what is used to determine ACCEPT or REJECT null hypothesis

Question 42

Mann-Whitney test

Answer 42

Ordinal data 2-groups Independent data compares MEDIANS

Question 43

Wilcoxon Signed Rank test

Answer 43

Ordinal data 2-group Paired/Related data compares MEDIANS

Question 44

Kruskal-Wallis test

Answer 44

Ordinal data ** 3+ groups ** Independent data compares MEDIANS ** If p value < 0.05, run a POST-HOC test to determine between which groups the difference exists!

Question 45

Friedman test

Answer 45

Ordinal data ** 3+ groups ** Paired/Related data compares MEDIANS **If p value < 0.05, run a POST-HOC test to determine between which groups the difference exists!

Question 46

Name Post-Hoc tests for ORDINAL data:

Answer 46

Student-Newman-Keul Dunnett Dunn **Kruskal-Wallis and Friedman**

Question 47

What type of tests utilize Post-Hoc tests?

Answer 47

Group Comparisons 3+ groups p value < 0.05 Ordinal data stat tests: - Kruskal-Wallis test - Friedman test Interval data stat tests: - ANOVA/ANCOVA - Repeated Measures ANOVA/ANCOVA

Question 48

(Pearson's) Chi-Square test

Answer 48

Nominal data 2-groups Independent data compares PROPORTIONS