SEM Flashcards

1
Q

What is SEM?

A

A confirmatory approach looking at the fit between observed data and a predefined model.

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

What is a latent variable?

A

An unobserved theoretical construct which is indicated by observed variables that load onto it.

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

What is an endogenous variable?

A

A latent variable influenced by another latent variable, or a DV.

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

What is an exogenous variable?

A

A latent variable that predicts fluctuations in another latent variable, an IV.

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

What is the focus of a measurement model?

A

The relationship between a latent variable and its measured variables.

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

What can CFA help us do?

A

Check that items contribute meaningfully to a factor, refine models and compare theoretical models.

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

What does Chi square assess?

A

The probability that the model fits the population.

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

What CFI indicates good fit?

A

> .95

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

What CFI indicates adequate fit?

A

> .90

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

What RMSEA indicates good fit?

A

< 0.05.

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

What RMSEA indicates adequate fit?

A

.05 - .08

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

What SRMR indicates good fit?

A

< 0.08.

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

What is the key extra step in a full latent variable model?

A

The researcher specifies the impact of one latent variable on another.

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

When interpreting latent variable relationships, what do standardised path coefficients convey?

A

A change in X number of SDs of endogenous variable is predicted by a change of 1 standard deviation in the exogenous variable.

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

For standardised path coefficients, what is a small effect size?

A

.05

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

For standardised path coefficients, what is a medium effect size?

A

.1

17
Q

For standardised path coefficients, what is a large effect size?

A

.25

18
Q

What values do you use to find out how much variance in an endogenous variable is accounted for?

A

Squared multiple correlations.