Chapter 20 Flashcards
Multilevel models should be used to analyse data that have a _____________structure.
hierarchical
If you are measuring depression after psychotherapy and your sample of patients will see different therapists within different clinics- how many levels are in this hierarchy?
This is a three-level hierarchy with:
- depression scores from patients
- nested within therapists
- who are nested within clinics
In ordinary regression, parameters generally are a ______ ______ estimated from the sample (a _____ ____)
In ordinary regression, parameters generally are a FIXED VALUE estimated from the sample (a FIXED EFFECT).
Hierarchical models are just like regression, except that you can allow… (this is called a ________effect)
Hierarchical models are just like regression, except that you can allow PARAMETERS TO VARY (this is called a RANDOM effect)
If we estimate a linear model within each context rather than a sample as a whole, then we can assume… or…
that the intercepts of these models vary OR that the slopes of these models differ
If intercepts of a model vary we have a …
random intercepts model
We can compare different multilevel models by looking at …
…the difference in the value of −2LL.
Under what circumstances would we compete different models by looking at the difference in the value of -2LL?
when we have changed one parameter or added one new thing to a model
For any model we have to assume a…
…covariance structure.