Multiple Regression Flashcards

1
Q

What are the aims of a multiple regression? (2)

A

a) Determine overall fit of a model

b) Establish relative contributions of the predictor variables

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What is the MR equation?

A

Y = a + b1 x1 + b2 x2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

In the MR equation (Y = a + b1 x1 + b2 x2 etc), what do a and b represent?

A
A = intercept 
B = slope
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

The variance in Y is explained by…(2)

A
  1. The IV

2. Residual error variance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

An MR assumes that IVs are ______ and ______ _______related to IV

A

Collectively
Independently
Linearly

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

4 assumptions of a multiple regression?

A

1) Linear relationship
2) Homoscedasticity
3) Normal distribution
4) Large sample size

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What values do we look at to determine HOW WELL THE MODEL FITS?

A

F
Df
P
R2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

The R2 tells us

A

The proportion of variance in the DV explained by the IVs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

“The proportion of variance in the DV explained by the IVs” What in the SPSS output tells us this?

A

R2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

The regression equation (SPSS output)…

A

F = (df), p =, R2

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What do the B values tell us in the coefficients table?

A

Change in the DV for every one unit of change in the IV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

“Change in the DV for every one unit of change in the IV” - what tells us this?

A

B weights

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

B weights may either be positive or negative to indicate…

A

The direction of change

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Which values do we look at in the coefficients table?

A

B Value,

Significance

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What should the table of coefficients include?

A

B, SEb & Beta

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Which 4 factors might we look at to determine the models overall fit?

A
  1. Multiple Correlation Coefficient
  2. Significance of overall model
  3. Percentage variance explained
  4. Precisions of predictions from model
17
Q

What are the three entry options of an MR?

A

Direct
Hierarchical
Stepwise

18
Q

Which entry option involves simultaneous entry?

A

Direct entry

19
Q

Which entry option involves a user-determined-order entry?

A

Hierarchical

20
Q

Hierarchical entry is usually based on…

21
Q

Hierarchical Entry

After IVs are added, the increase in…

A

R2 is examined

22
Q

Which entry option involves an SPSS-determined order?

23
Q

Which entry option comes with the most limitations?

24
Q

Direct Entry

How can we see the independent contributions?

A

Standardised beta and significance

25
2 Limitations of Step-wise entry?
(i) May capitalise on chance relationships | (ii) An important predictor may be excluded due to its correlation with another variable
26
Describe Hierarchical Entry
User-determined order based on theory
27
Describe step-wise entry
Discarded step-by-step according to statistical criteria | Highest correlating variable entered first
28
Step-wise entry Which variable is entered first?
Highest-correlating variable
29
You would use the entry to assess the effect of one group of predictors over and above that of another
Hierarchical
30
Hierarchical entry Who determines the order of entry?
Researcher
31
In all 3 types of MR, both ________ statistics and ________ statistics should be reported
Strength | Significance
32
Gender is a ____________ variable
Dichotomous
33
For HIERARCHICAL REGRESSION, when reporting the R2 CHANGE, do you report the F value from the ANOVA table?
No - model summary table