Quiz 4

Question 1

What are censored observations?

Answer 1

instances of incompleteness - Those who have not yet reached terminal event by end of the study

Question 2

What is hazard ratio?

Answer 2

compares how often a particular terminal event happens between 2 groups
- the difference in the steepness of the curve

Question 3

HR = 1

Answer 3

the event rates are the same in both groups

Question 4

HR > 1

Answer 4

event rate in the treatment group is faster than in the control group

Question 5

HR < 1

Answer 5

the event rate in the treatment group is slower than in the control group

Question 6

The _______ the HR is from 1, the larger the difference between the 2 groups

Answer 6

farther

Question 7

What is a log-rank test?

Answer 7

compares 2 or more samples with survival data in the presence of censored observations

Question 8

What is a Kaplan-Meier curve? Does it account for censored observations?

Answer 8

analyzes the probability of an event at specific time intervals

does account for censored observations

Question 9

Test can fail if the 2 curves _______

Answer 9

cross - there is no statistical power

Question 10

What are the limitations of log-rank test?

Answer 10

• can test only 1 variable at a time
• cannot control for confounders or other risk factors
• cannot include interaction times

Question 11

Which model can accommodate any number of confounding variables?

Answer 11

cox proportional hazard model (Cox Model)

Question 12

What assumptions are needed for exploratory factor analysis?

Answer 12

• continuous data
• normal distribution
• sample size is large enough (>200, >5 observations per variable)
• correlation is >0.3 between variables

Question 13

What are the limitations of exploratory factor analysis?

Answer 13

• subjective analysis
• variables may not always be generalizable to the population
• no causal interferences can be made

Question 14

What is a strong Eigenvalue?

Answer 14

greater than or equal to 1

Question 15

What are the criteria for the significance of factor loading?

Answer 15

• FL > 0.3: minimum considerations
• FL > 0.4: more important
• FL > 0.5: practically significant

Question 16

What are the limitations to confirmatory factor analysis?

Answer 16

• sample size required to be sufficiently large

- very sensitive to outliers and missing data

Question 17

What type of factor analysis will be run when there are outliers and missing data?

Answer 17

exploratory factor analysis

Question 18

What are the limitations of cluster analysis?

Answer 18

• method is high subjective (arbitrary decisions)

- easier to do with continuous data

Question 19

What is the interpretation range for a silhouette coefficient? Where are large and small coefficients?

Answer 19

• interpretation range is -1 to +1
• large coefficients are close to 1 - data is very well clustered
• small coefficients are close to 0 - data is not well clustered

Question 20

What are the good, fair, and poor ranges for silhouette coefficients?

Answer 20

Good: 0.5-1
Fair: 0.2-0.5
Poor: -1-0-0.2