603 Review Part 1

Week4

100

Ŷ = a + b₁X₁ + b₂X₂ + b₃X₃ is called a?

Regression equation

100

In MR, this value represents the unique relationship between Y and an X

Partial regression coeff.(or partial r2)

100

In MR, this is proportion of variability in Y explained by Xs as-a-set

R²

100

Which will tend to be smaller? Zero-order reg. wts or partial reg. wts?

Partial

100

In MR: Which represents variability acctd for by Xs? SSReg, SSResidual, SSTotal?

SSReg

Weeks6,7

100

Name for situation where Xs are highly correlated (+.70)

Multicollinearity

100

Variability in Y the same across levels of X

Homoskedasticity

100

What do we call a variable that comes between X and Y in a causal chain?

mediator

100

Another term for a mediated effect is?

An indirect effect

100

What do we call a variable that qualifies the conditions under which an effect holds?

Moderator

Weeks8,10

100

When testing for moderation in MR, if a sig. interaction is found, next steps?

Simple slopes

100

In MR, the interaction term has to be created by researcher. How is this done?

Multiplying the two predictors/Xs (X1*X2)

100

If [X1*X2] ns, next step?

Re-run MR without interaction term

100

Effect of stressors on health depends on social support. Social support is a

Moderator variable

100

One benefit of “centering” Xs in MR is that it reduces

Multicollinearity