Lecture 34-37: The Rest Of Biostats

Question 1

Nominal –> 2 Groups –> Independent

Answer 1

(Pearson’s) Chi-square test (X(squared))
Or
Fischer’s Exact

Question 2

Nominal –> 2 Groups –> Related

Answer 2

McNemar

Question 3

Nominal –> 3 or More Groups –> Independent

Answer 3

Chi-square test of Independence
Or
Fischer’s Exact

(2 of 3 Bonferonni’s Test of Inequality)

Question 4

Nominal –> 3 or More Groups –> Related

Answer 4

Cochran

One of Three Bonferronni Tests of Inferiority

Question 5

Nominal –> Proportion of Events (Survival)

Answer 5

Log-Rank

Question 6

Nominal –> Measure of Correlation –>

Answer 6

Contingency Coefficient

Question 7

Nominal –> Prediction or Association

Answer 7

Logistic Regression

Question 8

Ordinal –> 2 Groups –> Related

Answer 8

Wilcoxon-Signed-Rank

Question 9

Ordinal –> 2 Groups –> Independent

Answer 9

Mann Whitney

Question 10

Ordinal –> 3 or more groups –> Independent

Answer 10

Kruskal Wallis
-->
Student-Newman-Keul
Dunnett 
Dunn

Question 11

Ordinal –> 3 Or More Groups –> Related

Answer 11

Friedman

Post-Hoc:
Student-Newman-Keul
Dunnett
Dunn

Question 12

Ordinal –> Correlations

Answer 12

Spearman Correlation

Question 13

Ordinal –> Associations or Predictors

Answer 13

Multinomial Logistic Regression

Question 14

Ordinal –> Survival

Answer 14

Cox Proportional Hazard

Question 15

Interval –> 2 Groups –> Independent

Answer 15

Student t

Question 16

Interval –> 2 Groups –> Related

Answer 16

Paired t

Question 17

Interval –> 3 Or More Groups –> Independent

Answer 17

ANOVA or MANOVA

Leads to Confounders Present

Leads to ANCOVA or MANCOVA

Question 18

Interval –> 3 or More Groups –> Related

Answer 18

Repeated Measures ANOVA
Or
Repeated Measures MANCOVA

If Confounder Present:

Repeated Measures ANCOVA
Or
Repeated Measures MACNOVA

Question 19

Interval –> Survival

Answer 19

Kaplan-Meier

Question 20

Interval –> Correlation

Answer 20

Pearson Correlation

Question 21

Interval –> Prediction or Association

Answer 21

Linear Regression

Question 22

Name the 6 Post-Hoc Tests

Answer 22

Bonferonni
Tukey
Scheffe
Dunn
Dunnett
Student-Newman-Keul

Question 23

When Determining the Correct Statistical Test (and after deciding what data type it is), what is the next question to ask to determine which test to use?

Answer 23

What Type of Comparison/Assessment is desired?

Question 24

Define Correlations

Answer 24

• Correlation (r)
• Provides a QUANTITATIVE measure of the STRENGTH & DIRECTION of a relationship between variables
• -VALUES RANGE FROM -1.0 TO +1.0
• Partial Correlation
• A correlation that controls for confounding variables

Question 25

What are the three types of correlation tests?

Answer 25

* Nominal Correlation test = CONTINGENCY COEFFICIENT * Ordinal Correlation test = SPEARMAN CORRELATION * Interval Correlation test = PEARSON CORRELATION – p>0.05 for a Pearson Correlation just means there is no LINEAR correlation; there may still be NON-LINEAR correlations present! - ALL CORRELATIONS CAN BE RUNS AS A "PARTIAL CORRELATION" TO CONTROL FOR CONFOUNDING

Question 26

Describe Survival Tests

Answer 26

* Compares the proportion of, or time-to, event occurrences between groups - Commonly represented by a KAPLAN-MEIER CURVE

Question 27

List the types of survival Tests

Answer 27

Event-Occurrence / Time-to-Event --> SURVIVAL TEST – Nominal Survival test = LOG-RANK TEST – Ordinal Survival test = COX-PROPORTIONAL HAZARDS TEST – Interval Survival test = KAPLAN-MEIER TEST -ALL CAN BE REPRESENTED BY A KAPLAN-MEIER CURVE

Question 28

Define A Regression Test

Answer 28

* Provide a measure of the relationship between variables by allowing the prediction about the dependent, or outcome, variable (DV) knowing the value/category of independent variables (IV’s) - ALSO ABLE TO CALCULATE ODDS RATIO FOR A MEASURE OF ASSOCIATION

Question 29

List the Possible Regression Tests

Answer 29

Outcome Prediction/Association (OR) --> REGRESSION – Nominal Regression test = LOGISTIC REGRESSION – Ordinal Regression test = MULTINOMIAL LOGISTIC REGRESSION – Interval Regression test = LINEAR REGRESSION

Question 30

Once again, what are the 4 big questions to ask regarding choosing the correct statistical test

Answer 30

1. What DATA LEVEL is being recorded? a. Does the data have MAGNITUDE? (yes/no) b. Does the data have a fixed, measureable INTERVAL along the entire scale? (yes/no) 2. What TYPE OF COMPARISON/ASSESSMENT is desired? – Frequencies/Counts/Proportions* 3. *HOW MANY GROUPS are being compared? - 2 or 3 or more groups 4. *Is the data INDEPENDENT or RELATED (PAIRED)? - Data from the same (paired) or different groups (independent)

Question 31

With Three or More Groups of Independent Nominal Data, Describe the test used.

Answer 31

* Chi-square test of Independence (X2) - Both this test and the Pearson Chi-Square test for 2 Groups compares group proportions and if they are different from that expected by chance * Assumptions: - Usual chi-square (binomial) distribution for nominal-type data - No cell with Expected count of <5 * For ≥2 Groups with EXPECTED cell count of <5: - Use Fisher’s Exact test * For statistically significant findings (p<0.05) in 3 or more comparisons, one Must perform subsequent analysis (POST-HOC TESTING) to determine which groups are different: - Multiple X2 tests are NEVER acceptable - - Risk of Type 1 error increases with each additional test! (almost guaranteed after 4-5 tests) - BONFERRONI TEST OF INEQUALITY (BONFERRONI CORRECTION) - - Adjusts the p value for # of comparisons being made - - Very conservative

Question 32

Describe Paired/Related Nominal Data

Answer 32

* 2 Groups of Paired/Related Data* - MCNEMAR test * ≥3 Groups of Paired/Related Data* - COCHRAN - - Same principle and assumptions as X2 yet mathematically factors in concept of paired, or related, data - - BONFERRONI TEST OF INEQUALITY (BONFERRONI CORRECTION): Adjusts the p value for # of comparisons being made. Very conservative. * KEY WORDS FOR “PAIRED” or “RELATED” DATA: “Pre- vs. Post-”, “Before vs. After”, “Baseline vs. End”, etc…

Question 33

Describe Independent Ordinal Data

Answer 33

* 2 Groups of Independent Data - MANN-WHITNEY TEST * ≥3 Groups of Independent Data - KRUSKAL-WALLIS TEST - - Both tests compares the median values between groups. Both also used for Interval data not meeting parametric requirements - - If 3+ group comparison significant, must perform a POST-HOC TEST to determine where difference(s) is(are)…

Question 34

Describe Paired/Related Ordinal Data

Answer 34

* 2 Groups of Paired/Related Data* - WILCOXON SIGNED RANK TEST * ≥3 Groups of Paired/Related Data* - FRIEDMAN TEST - - Both tests compares the median values between groups. Each also effective for non-normally distributed Interval data or don’t meet all parametric requirements - - If 3+ group comparison significant, must perform a POST-HOC test to determine where differences are… * KEY WORDS FOR “PAIRED” or “RELATED” DATA: “Pre- vs. Post-”, “Before vs. After”, “Baseline vs. End”, etc…

Question 35

Describe the Post-Hoc Tests for Ordinal Data involving 3 or more groups

Answer 35

* STUDENT-NEWMAN-KEUL - Compares all pairwise comparisons possible - All groups must be equal in size * DUNNETT test - Compares all pairwise comparisons against a Single Control - All groups must be equal in size * DUNN test - Compares all pairwise comparisons possible - Useful when all groups are Not of equal size

Question 36

Incomplete

Answer 36

Resume on Slide 94