6 Structural Equation Modelling Flashcards

1
Q

What is the aim of path analysis?

A

To estimate the magnitude and significance of hypothesised causal connections between variables.

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

What is the path coefficient for a residual term? And why?

A

Always 1. It is assumed that error is not systematically associated with anything else.

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

In path analysis, what is disturbance?

A

Error/residual

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

What are endogenous variables? How do they work with arrows?

A

Presumed effects. Have arrows going into them.

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

What are exogenous variables? How do they work with arrows?

A

Presumed causes. Variables WITHOUT arrows going into them. Arrows go from them to other causes.

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

What is a path coefficient?

A

Value of the arrow/estimate of cause.

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

How many regression equations per endogenous variable do you need in a path model?

A

1

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

How do you calculate indirect effects?

A

By multiplying the path coefficients.

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

How do you calculate total effects?

A

Adding direct and indirect effects.

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

What’s the difference between recursive and non-recursive models?

A

Recursive – all paths go in one direction with no feedback loops. Can be solved with standard multiple regression equations.

Non-recursive –have reciprocal causality/feedback loops. Cannot be solved with regression equations.

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

What’s the difference between:

1) just-identified
2) under-identified
3) over-identified models?

A

1) Under-identified models have less than one possible solutions (i.e. they cannot be solved).
2) Just-identified – there is only one solution to what paths are.
3) Over-identified –have more than one solution. SEM programs determine which is most likely (e.g. CFA models which are rotated for best fit).

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

A just-identified (saturated) model has __ ___ elements in the covariance matrix as there are ________ __ __ _______.

A

A just-identified (saturated) model has as many elements in the covariance matrix as there are parameters to be estimated.

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

In an underidentified model there are more or less elements in the correlation matrix than there are parameters to be estimated?

A

Underidentified - less elements in the correlation matrix than there are parameters to be estimated.

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

How do you calculate the number of elements in the correlation matrix if the number of variables is greater than 3?

A

Use the formula for triangular numbers.

K= (N(N+1)) / 2

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

In an underestimated model in SEM, df is _______.

A

In an underestimated model in SEM, df is negative.

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

In an overidentified model, KNOWNS > UNKNOWNS. True or false?

A

True

17
Q

What are regarded as the ‘knowns’ and ‘unknowns’ in model identification?

A

Knowns –elements in correlation matrix

Unknowns –parameters to be estimated (e.g. coefficients)

18
Q

For a model to be identified it must have how many degrees of freedom?

A

0 or more

19
Q

How do you calculate df in a model?

A

Number of parameters to be estimated

20
Q

In an AMOS Regression Weight output, what is the Critical Ratio?

A

The estimate (unstandardized correlation coefficient) divided by standard error.

21
Q

What are squared multiple correlations?

A

The same as R-squared. They represent the proportion of variance in endogenous variables that the model explains.

22
Q

What is the logic of calculating model fit?

A

Estimate covariance matrix if model was 100% accurate in describing data. Compare this to actual covariance matrix. Calculate discrepancy with chi-square.

23
Q

How can chi-square estimates be biased in SEM?

A

By having large sample –> significant chi-square.

24
Q

What does a significant chi-square mean?

A

The model does NOT describe the data well. There is a large discrepancy between model and data.

25
Q

What values would you want to find in the:
CFA (confirmatory fit index)
GFI (goodness of fit index)
RMSEA (root mean square error of approximation)

A

CFI and GFI –values over .90

RMSEA –values less than .6

26
Q

If chi-square is less than df, what is this, what does it mean?

A

It’s called overfit, and it means that your model is too complex. Also results in maxing out GFI, NFI, CFI and RMSEA.