WEEK 8

Question 1

What is a covariate and what is its purpose in a model

Answer 1

A variable that might affect the results of your study and is included in the analysis to control for its influence.
Reduces unexplained variation in our response and allows us to isolate the effect of the predictor that we are interested in.
Reduces confounding and increases accuracy

Question 2

How do we graph a model where there is one continuous predictor as a covariate and a categorical predictor

Answer 2

Add separate lines in the graph accounting for the levels of the categorical predictor

Question 3

What is the purpose of an analysis of covariance

Answer 3

Allows us to attribute variation in our response variable to different sources, which allows us to test the hypothesis about the effect of each predictor

Question 4

How can we tell if there is an interaction present between predictor and covariate

Answer 4

By analysing the graph, if slopes are parallel, it means there are no significant interaction.
If slopes are not parallel, there is.

Question 5

is it ok to interpret the main effect of a predictor when the’re evidence of an interaction between categ and cont predictors?

Answer 5

No. We must describe how the effect of one predictor depends on the level of the other

Question 6

What are the basic assumptions of ANCOVA?

Answer 6

Same as ANOVA:
Data must be independent
Relationship must be linear
residuals must have equal variance
residuals are normally distributed
BUT with one extra:
range of continuous predictor (covariate) must be similar for all levels of the categorical predictor (lines should overlap along x-axis)