Lecture 34-37: The Rest Of Biostats Flashcards
Nominal –> 2 Groups –> Independent
(Pearson’s) Chi-square test (X(squared))
Or
Fischer’s Exact
Nominal –> 2 Groups –> Related
McNemar
Nominal –> 3 or More Groups –> Independent
Chi-square test of Independence
Or
Fischer’s Exact
(2 of 3 Bonferonni’s Test of Inequality)
Nominal –> 3 or More Groups –> Related
Cochran
One of Three Bonferronni Tests of Inferiority
Nominal –> Proportion of Events (Survival)
Log-Rank
Nominal –> Measure of Correlation –>
Contingency Coefficient
Nominal –> Prediction or Association
Logistic Regression
Ordinal –> 2 Groups –> Related
Wilcoxon-Signed-Rank
Ordinal –> 2 Groups –> Independent
Mann Whitney
Ordinal –> 3 or more groups –> Independent
Kruskal Wallis --> Student-Newman-Keul Dunnett Dunn
Ordinal –> 3 Or More Groups –> Related
Friedman
Post-Hoc:
Student-Newman-Keul
Dunnett
Dunn
Ordinal –> Correlations
Spearman Correlation
Ordinal –> Associations or Predictors
Multinomial Logistic Regression
Ordinal –> Survival
Cox Proportional Hazard
Interval –> 2 Groups –> Independent
Student t
Interval –> 2 Groups –> Related
Paired t
Interval –> 3 Or More Groups –> Independent
ANOVA or MANOVA
Leads to Confounders Present
Leads to ANCOVA or MANCOVA
Interval –> 3 or More Groups –> Related
Repeated Measures ANOVA
Or
Repeated Measures MANCOVA
If Confounder Present:
Repeated Measures ANCOVA
Or
Repeated Measures MACNOVA
Interval –> Survival
Kaplan-Meier
Interval –> Correlation
Pearson Correlation
Interval –> Prediction or Association
Linear Regression
Name the 6 Post-Hoc Tests
Bonferonni Tukey Scheffe Dunn Dunnett Student-Newman-Keul
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?
What Type of Comparison/Assessment is desired?
Define Correlations
- 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
What are the three types of correlation tests?
- 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
Describe Survival Tests
- Compares the proportion of, or time-to, event occurrences between groups
- Commonly represented by a KAPLAN-MEIER CURVE
List the types of survival Tests
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
Define A Regression Test
- 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
List the Possible Regression Tests
Outcome Prediction/Association (OR) –> REGRESSION
– Nominal Regression test = LOGISTIC REGRESSION
– Ordinal Regression test = MULTINOMIAL LOGISTIC REGRESSION
– Interval Regression test = LINEAR REGRESSION
Once again, what are the 4 big questions to ask regarding choosing the correct statistical test
- 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) - What TYPE OF COMPARISON/ASSESSMENT is desired?
– Frequencies/Counts/Proportions* - *HOW MANY GROUPS are being compared?
- 2 or 3 or more groups - *Is the data INDEPENDENT or RELATED (PAIRED)?
- Data from the same (paired) or different groups (independent)
With Three or More Groups of Independent Nominal Data, Describe the test used.
- 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
Describe Paired/Related Nominal Data
- 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…
Describe Independent Ordinal Data
- 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)…
Describe Paired/Related Ordinal Data
- 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…
Describe the Post-Hoc Tests for Ordinal Data involving 3 or more groups
- 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
Incomplete
Resume on Slide 94
