Structural Equation Modelling

Question 1

SEM

Answer 1

A very flexible technique that uses latent variables in order to answer complex questions. SEM develops a model then determines if the data could have been generated using the created model.

Question 2

Latent Variable

Answer 2

Assumed to exist from shared variance between observable indicators. Assumed to be error free because it is the shared variance between all measured variables.

Question 3

Underidentified Model

Answer 3

Too many unknowns in the model. More parameters than data points.

Question 4

Parameter

Answer 4

Set of values that indicate the nature of the relationship between variables. Can either be fixed (not from data) or free (from data)

Question 5

Just Identified Model

Answer 5

Number of parameters equals the number of data points. Uses all information so cannot test model.

Question 6

Over Identified Model

Answer 6

Number of parameters exceeds number of data points. Not all information is used so model can be tested and compared.

Question 7

Saturated Model

Answer 7

Contains all possible pathways. Perfect fit but very complex.

Question 8

Independence Model

Answer 8

All pathways are assumed to be zero. Very simple but basically useless.

Question 9

Default Model

Answer 9

Hypothesised model.

Question 10

Fit Indices

Answer 10

Range of statistics. Look at multiple indices to determine if the model is a good fit. Typically use CMIN and RMSEA (want below .08)

Question 11

Assumptions of SEM

Answer 11

Requires large sample, normality, linearity, no outliers, must be based on theory (confirmatory)

Question 12

Structural Model

Answer 12

Defines the relationship between latent and measured variables.

Question 13

Measurement Model

Answer 13

Defines the latent variables (what makes up the latent variables)

Question 14

Endogenous Variables

Answer 14

Receive and exert directional influence on other variables in the model

Question 15

Exogenous Variables

Answer 15

Don’t receive or exert influence on other variables.

Question 16

Model Fit in SEM

Answer 16

Model fit test assesses whether the model could have generated the data. If significant, it means that the model and the data are significantly different, meaning that the model is a poor one.