Lesson 9: Other Tests

Question 1

Logistic Regression

Answer 1

A regression model that measures the relationship between a categorical dependent variable and one or more independent variables

Question 2

Multilevel data

Answer 2

Data that represents more than one level, in which the lower level is nested within a higher level

Question 3

Multilevel Model

Answer 3

statistical approach to analyzing data that represents multiple levels

Question 4

Survival Analysis

Answer 4

A statistical analysis that models time to event data

Question 5

What is a logistic regression model used for?

Answer 5

predicting the outcome of a categorical dependent variable based on one or more predictor variables

Question 6

How is logistic regression different than one-way ANOVA and independent t-test?

Answer 6

logistic regression can predict the probability of the outcome variable

Question 7

How is logistic regression different than linear regression?

Answer 7

Linear regression predicts an OUTCOME based on one or more independent variables – logistic regression predicts the PROBABILITY of the outcome based on one or more independent variables.

Question 8

Results for independent variables in a logistic regression model are presented as?

Answer 8

odds ratios.

Question 9

Nested Data

Answer 9

-used when characteristics of more than one level are believed to be important

-

Question 10

Survival Analysis examines what sort of data?

Answer 10

time-to-event data

the time duration until one more events happen

• subjects in these studies are followed over a specific time period with special attention to the time at which the event of interest occurs
  ex: occurance of disease, death,

Question 11

What does logistic regression predict?

Answer 11

The probability of an outcome ;; Logistic regression measures the relationship between a categorical dependent variable and one or more independent variable to provide the probability of an outcome.

Question 12

Data is collected to examine the relationship between hospitals (n = 30 hospitals), doctor prescribing behaviors (n = 25 doctors per hospital), and patients (n = 50 patients per doctor). Why are ANOVA, t-tests, and basic regression analyses not appropriate for analyzing all of this data?

Answer 12

Nested data yields clusters of observations that cannot be considered independent – In other words, a cluster of patients seeing a particular doctor are likely to have similar experiences to one another and these experiences are likely to differ from those of a cluster of patients seeing a different doctor.

Question 13

T/F: Survival analysis has a dichotomous outcome.

T/F: Cox proportional hazards models can be used for interpreting results.

T/F: Kaplan-Meier curves are commonly used to illustrate time to event.

Answer 13

True

True

True