Deck 22: Structural Equation Modeling and Path Analysis

Question 1

SEM is a procedure for estimating a series of dependence relationships among a set of concepts or constructs represented by multiple measured variables and incorporated in to an integrated model.

Answer 1

True

Question 2

Dependence relationships are represented with one-headed straight arrows in a path model.

Answer 2

True

Question 3

SEM models are estimated so as to minimize the difference between the observed and
estimated covariance matrices.

Answer 3

True

Question 4

SEM analysis is based on raw data,rather than a correlation or covariance.

Question 5

SEM models are estimated so as to minimize the sum of squares between constructs.

Question 6

When two constructs are theoretically unrelated to each other,the path between those two
constructs should be be assigned a value of 1.

Question 7

When two constructs are theoretically unrelated to each other,the path between those two
constructs should be "free."

Question 8

Having at least three observed variables for each latent construct is a way to help make sure
that the model is identified.

Question 9

When two constructs are theoretically unrelated to each other,the path between those two
constructs should be "fixed."

Question 10

Measured variables or indicators are the actual items that are measured using a survey
questionnaire.

Question 11

It is recommended that each construct be measured by using at least three observed variables.

Question 12

Dependence relationships are represented with two-headed curved arrows in a path model.

Question 13

The confirmatory factor analysis is used to verify the factor structure of a set of observed
variables.

Question 14

An exogenous construct is represented in an structural path model as a an oval with no
one-headed arrows going into it.

Question 15

An exogenous construct is represented in an structural path model as a box with one-headed arrows going into it.

Question 16

A construct is an unobservable or latent variable that can be defined in conceptual terms that
can be measured directly.

Question 17

Generally,we use SEM to find a suitable model.

Question 18

Constructs are the actual items that are measured using a survey questionnaire.

Question 19

SEM models are estimated so as to maximize the likelihood function.

Question 20

SEM is distinguished by the explaining of the covariance among the measured items or
observed variables.

Question 21

A 0.90 value for RMSEA is considered conservative.

Question 22

In moving from the measurement model to the structural model the endogenous constructs
become exogenous.

Question 23

TLI is an absolute fit index.

Question 24

NFI is an incremental fit index.

Question 25

The baseline model that is commonly used is the null model that is based on the assumption
that the observed variables are uncorrelated.

Question 26

RNI is a parsimony fit index.

Question 27

In moving from the measurement model to the structural model the error terms are correlated
with each other.

Question 28

The impact of the average error variance of indicators can be understood in terms of communality.

Question 29

CFI is an absolute fit index.

Question 30

A 0.07 value for RMSEA is considered conservative.

Question 31

GFI of .95 or higher is a way to help make sure that the model is identified.

Question 32

In moving from the measurement model to the structural model the emphasis shifts from the
relationships between latent constructs and measured variables to the nature and magnitude of the relationships between constructs.

Question 33

RMSEA is an incremental fit index.

Question 34

A 0.04 value for RMSEA is considered conservative.

Question 35

Convergent validity is established when the measured items are conceptually consistent with a
construct definition.

Question 36

Construct validity is established when the measured items are conceptually consistent with a construct definition.

Question 37

CFI is an incremental fit index.

Question 38

RMSR is an absolute fit index.

Question 39

In moving from the measurement model to the structural model the exogenous constructs each
become endogenous.

Question 40

TFI is a fit measure used in SEM

Question 41

Δχ2 statistic is most useful for comparing nested models.

Question 42

SEM can be thought of as a combination of MANOVA and Exploratory factor analysis.

Question 43

SEM can be thought of as a combination of MANOVA and ANOVA.

Question 44

The specification of which variables load on which factors is a difference contrasting EFA with CFA.

Question 45

SEM has the following characteristics:

A) representation of constructs as unobservable or latent factors in dependence relationships.
B) estimation of multiple and interrelated dependence relationships incorporated in an integrated model.
C) explanation of covariance among the observed variables.
D) incorporation of measurement error in an explicit manner.
E) All of the above

Question 46

SEM can be thought of as a combination of Discriminant analysis and cluster analysis.

Question 47

SEM can be thought of as a combination of MANOVA and multiple regression analysis.

Question 48

Path analysis can be viewed as a special case of SEM.

Question 49

Which of the following are true regarding exogenous constructs?

A) An exogenous construct is the latent,multi-item equivalent of an independent variable in traditional multivariate analysis.
B) Graphically,an exogenous construct does not have any paths coming to it.
C) An exogenous construct is determined by factors outside of the model.
D) A and C are correct.
E) A,B,and C are correct.

Question 50

RMSR statistic is most useful for comparing nested models.

Question 51

Once the validity of the measurement model has been established,you can proceed with the
specification of structural model.

Question 52

A proposed structural model involves three constructs,one of which are exogenous.X1 causes Y1,which in turn causes Y2.This model is nonrecursive.

Question 53

SEM can be thought of as a combination of Factor analysis and multiple regression analysis.

Question 54

The size of the loading estimates considered significant is a difference contrasting EFA with CFA.

Question 55

A sample size of 1000 is used to test a structural model of advertising effectiveness for the
Coca-Cola company.The results show that a significant relationship exists between advertising effectiveness and sales.The estimated value of the standardized path estimate is .035.The Coca-Cola management dismisses the finding because the relationship is trivial.

Question 56

Which of the following is true about Structural Equation Modeling?

A) SEM explicitly takes into account the measurement error.
B) In SEM,the constructs are unobservable or latent factors that are representative by multiple variables.
C) SEM is mainly used as an exploratory technique.
D) A and B are true
E) A,B,and C are true.

Question 57

PNFI statistic is most useful for comparing nested models.

Question 58

________ are the actual items that are measured using a survey questionnaire.

A) Constructs
B) Observed factors
C) Measured variables or indicators
D) Estimated Parameter
E) Latent factors

Question 59

In SEM,the posited relationships between constructs are tested similarly to the way they are tested using cluster analysis.

Question 60

A proposed structural model involves four constructs,two of which are exogenous.X1 and
X2 together cause Y1,which in turn causes Y2.This model is recursive.

Question 61

Which is not an absolute fit index?

A) Chi-square
B) SRMR
C) NFI
D) CFI
E) A and C

Question 62

Which of the following are true regarding SEM models?

A) A structural model shows how the constructs are interrelated to each other,often with multiple dependence relationships.
B) A SEM model consists of two models: the measurement model and the structural model.
C) The measurement model depicts how the observed variable represent constructs.
D) The observed variables are measured by researcher.
E) All of the above.

Question 63

Dependence relationships are represented with ________ in a path model.

A) two-headed curved arrows
B) boxes
C) two-headed straight arrows
D) ovals
E) None of the above

Question 64

Model fit includes the following steps:

A) Estimate the covariance matrix between the observed variables based on the proposed measurement and structural models.
B) Model fit is determined by comparing how closely the estimated covariance matrix matches the observed covariance matrix.
C) Produce correlations or covariances among the items.
D) Examine the residuals.
E) All of the above.

Question 65

Which is not an incremental fit index?

A) GFI
B) RMSR
C) CFI
D) NFI
E) A and D

Question 66

Which of the following is not a fit measure?

A) TFI
B) CFI
C) GFI
D) RMSR
E) SRMR

Question 67

How is an exogenous construct represented in an structural path model?

A) It is represented by a box with one-headed arrows going into it and two headed arrows.
B) The construct is represented by an oval with no one-headed arrows going into it.
C) It is represented by an oval with no arrows.
D) It is represented by a box with one-headed arrows going into it.
E) None of the above

Question 68

The diagnostic cues that can be used to make appropriate modifications include:

A) specification search.
B) modification indices.
C) standardized residual.
D) the path estimates or loadings.
E) All of the above

Question 69

SEM can be thought of as a combination of which two techniques?

A) MANOVA and multiple regression analysis
B) MANOVA and ANOVA
C) MANOVA and Exploratory factor analysis
D) Discriminant analysis and cluster analysis
E) Factor analysis and multiple regression analysis

Question 70

________ is established when the measured items are conceptually consistent with a construct definition.

A) Discriminant validity
B) Nomological validity
C) Convergent validity
D) Composite reliability
E) None of the above

Question 71

When two constructs are theoretically unrelated to each other,the path between those two constructs should be ________.

A) free
B) assigned a value of 1
C) fixed
D) estimated
E) None of the above

Question 72

Which statistic is most useful for comparing nested models?

A) RMSR
B) PNFI
C) AGFI
D) Δχ2
E) NNFI

Question 73

A sample size of 1000 is used to test a structural model of advertising effectiveness for the Coca-Cola company.The results show that a significant relationship exists between advertising effectiveness and sales.The estimated value of the standardized path estimate is .035.The Coca-Cola management dismisses the finding.Why?

A) The sample size is probably too large.
B) The sample size is probably too small.
C) The relationship is trivial.
D) The model fit is poor.
E) None of the above

Question 74

In moving from the measurement model to the structural model:

A) The endogenous constructs become exogenous.
B) The emphasis shifts from the relationships between latent constructs and measured variables to the nature and magnitude of the relationships between constructs.
C) The exogenous constructs each become endogenous.
D) The error terms are correlated with each other.
E) None of the above

Question 75

SEM models are estimated so as to:

A) maximize the likelihood function.
B) A and C
C) minimize the sum of squares between constructs.
D) minimize the difference between the observed and estimated covariance matrices.
E) minimize the least squared difference.

Question 76

The validity of the measurement model depends on:

A) Convergent validity.
B) Discriminant validity.
C) Goodness-of-fit results.
D) Composite reliability.
E) All of the above

Question 77

Which of the following steps are involved in conducting SEM?

A) Assess measurement model reliability and validity.
B) Define the individual constructs.
C) Specify the measurement model.
D) Specify the structural model if the measurement model is valid.
E) All of the above.

Question 78

A proposed structural model involves four constructs,two of which are exogenous.X1 and X2 together cause Y1,which in turn causes Y2.What term describes this model?

A) descriptive
B) nonrecursive
C) empirical
D) recursive
E) None of the above

Question 79

In a tests of a structural model,the path between store preference and store patronage has a coefficient of .85.What does this provide evidence of?

A) the hypothesis that preference causes store patronage is supported
B) absence of other casual factors
C) time order of occurrence
D) nonspurious association
E) None of the above

Question 80

Which is a way to help make sure that the model is identified?

A) have at least three observed variables for each latent construct
B) none of the above
C) a large sample size
D) have all formative indicators
E) a GFI of .95 or higher