Deck 15: Other Multivariate Techiques

Question 1

Statistical techniques that do not designate variables as either dependent or independent are performed for the purpose of _____.

A) testing hypotheses
B) collecting secondary data
C) determining covariation
D) data reduction
E) determining causality

Answer 1

D

Question 2

_____ techniques divide variables into independent and dependent sets for analysis purposes.

A) Dependence
B) Interdependence
C) Factoring
D) Discriminating
E) Multivariate

Answer 2

A

Question 3

_____ techniques use only one set of variables at a time.

A) Dependence
B) Interdependence
C) Discriminating
D) Multivariate
E) Univariate

Answer 3

B

Question 4

This multivariate statistical technique can summarize the information from a large number of variables into a much smaller number of variables by identifying latent relationships and combining variables into a few factors. This technique is called _____.

A) Cluster analysis
B) Regression analysis
C) Factor analysis
D) ANOVA
E) Discriminant analysis

Question 5

Carole has 50 variables on a questionnaire and would like to reduce that down to about five that explain over 60% of the variance in the data. Carole should use which multivariate statistical technique?

A) regression analysis
B) ANOVA
C) correlation analysis
D) factor analysis
E) cluster analysis

Question 6

Factor analysis is a statistical technique that develops _____ combinations of variables that summarize the original variables based on their underlying patterns.

A) curvilinear
B) linear
C) nonlinear
D) normal
E) logarithmic

Question 7

When using factor analysis, a linear combination of the original variables is called a(n) _____.

A) factor
B) function
C) weight
D) coefficient
E) variance

Question 8

The first factor derived in factor analysis is the combination of variables that accounts for _____ of the variance in the data as a whole than any other combination of the original variables.

A) less
B) more
C) 20%
D) 60%
E) 90%

Question 9

What are the two basic factor models?

A) principle component analysis and common factor analysis
B) hierarchical and nonhierarchical
C) agglomerative and divisive
D) demographic and centroid
E) simple and complex

Question 10

_____ analysis uses all of the variance in the data set when performing factor analysis.

A) K-means
B) Hierarchical
C) Nonhierarchical
D) Principle components
E) Common factor

Question 11

_____ analysis is based only on the common variance when performing factor analysis.

A) K-means
B) Hierarchical
C) Nonhierarchical
D) Principle components
E) Common factor

Question 12

Total variance can be divided into all of the following types of variance EXCEPT _____.

A) error
B) common
C) unique
D) linear
E) all of these choices are types of variance that make up total variance

Question 13

The _____ variance is that portion of the total variance that is shared with all the original variables in the analysis.

A) error
B) common
C) unique
D) linear
E) factor

Question 14

The _____ variance is that portion of the total variance that is specific to only one variable.

A) error
B) common
C) unique
D) singular
E) univariate

Question 15

Laurie knows that the unique and error variances are a relatively small portion of the total variance and wants to reduce the original set of variables into a smaller set of composite variables. Which type of multivariate analysis technique would be most appropriate for Laurie to use?

A) common factor analysis
B) k-means cluster analysis
C) principle components factor analysis
D) regression analysis
E) ANOVA

Question 16

Hugo knows little about the unique and error variance and he wants to examine the original variables in terms of their common underlying dimensions. Which type of multivariate statistical technique should he use?

A) common factor analysis
B) k-means cluster analysis
C) principle components factor analysis
D) regression
E) ANOVA

Question 17

What are the two options for factor rotation?

A) obtuse and oblique
B) hierarchical and nonhierarchical
C) principle components and common factor
D) orthogonal and oblique
E) agglomerative and divisive

Question 18

If a researcher desires to rotate a factor analysis solution so that the factors are independent of one another and the correlation between them is 0, which factor rotation solution should be used?

A) perpendicular
B) orthogonal
C) oblique
D) parallel
E) complete

Question 19

If a researcher desires to rotate a factor analysis solution that permits the derived factors to be correlated with one another, which factor rotation solution should be used?

A) perpendicular
B) orthogonal
C) oblique
D) parallel
E) complete

Question 20

The term _____ is used to describe factor analysis results in which each original variable has a high loading on only one factor and relatively low loadings on the other derived factors.

A) factor rotation
B) oblique
C) orthogonal
D) nonhierarchical
E) simple structure

Question 21

Which of the following represents the correlation between an original variable and a derived factor?

A) factor loading
B) factor weight
C) factor coefficient
D) α
E) β

Question 22

Two of the most widely used methods to decide the number of factors to retain are the _____.

A) hierarchical and nonhierarchical criteria
B) dependence and interdependent criteria
C) oblique and orthogonal criteria
D) latent root and percentage of variance criteria
E) loading and varimax criteria

Question 23

The latent root criterion for determining the number of factors to retain is that with principle components analysis, factors that have a latent root of _____ or higher are retained.

A) 1
B) 2
C) 5
D) 10
E) 15

Question 24

The rule of thumb for the percentage of variance criterion for retaining factors in factor analysis is that a factor solution should account for a minimum of _____ percent of the total variance.

A) 10
B) 20
C) 50
D) 60
E) 80

Question 25

The _____ is the total amount of variance in a factor analysis.

A) trace
B) eigenvalue
C) communality
D) Z score
E) loading

Question 26

_____ represents the row sum of squared factor loadings and tells us how much of the variance in a particular variable is accounted for by the factor solution.

A) Eigenvalue
B) Communality
C) Hit ratio
D) Trace
E) Z score

Question 27

_____ refers to a variable loading on more than one factor.

A) Orthogonal loading
B) Oblique loading
C) Hierarchical loading
D) Communality
E) Cross loading

Question 28

A factor analysis solution that does not show the smaller, less significant factor loadings is called a(n) _____.

A) contingency table
B) easy-read matrix
C) orthogonal rotation
D) communal output
E) simple structure analysis

Question 29

_____ is an interdependence multivariate technique that enables us to combine objects into groups so that objects within each of the groups are similar to each other and different from objects in all other groups.

A) Factor analysis
B) Regression analysis
C) Cluster analysis
D) Discriminant analysis
E) Grouping analysis

Question 30

Which is NOT a phase to be performed in cluster analysis?

A) Divide that total sample into smaller subgroups.
B) Verify the groups are statistically different and theoretically meaningful.
C) Profile the clusters by describing the characteristics of each cluster in terms of demographics, psychographics, or other relevant characteristics.
D) Develop a linear function and use it to predict in which group a survey respondent belongs.
E) All of these choices are phases in cluster analysis.

Question 31

The distances between the objects we are clustering can be measured in several ways. Which measure is most commonly used and has the fewest weaknesses?

A) squared Euclidean distance
B) sum of the absolute differences
C) Mahalanobis distance
D) Hierarchical distance
E) Nonhierarchical distance

Question 32

Procedures used to group objects that are similar in cluster analysis all can be classified into two types, which are ____.

A) dependence and interdependent
B) hierarchical and nonhierarchical
C) orthogonal and oblique
D) agglomerative and divisive
E) principle components and common

Question 33

Agglomerative and divisive approaches to cluster analysis are types of _____ clustering procedures.

A) hierarchical
B) nonhierarchical
C) orthogonal
D) oblique
E) dependence

Question 34

A tree-graph for hierarchical cluster analysis is referred to as a(n) _____.

A) factor matrix
B) discriminant function
C) dendogram
D) eigen structure
E) box and whisker plot

Question 35

In _____ clustering procedures, one or more cluster seeds (initial starting points) are selected and objects within a prespecified distance from the cluster seeds are considered to be in particular cluster.

A) factor
B) orthogonal
C) oblique
D) hierarchical
E) nonhierarchical

Question 36

What is an advantage of the nonhierarchical clustering over the hierarchical clustering procedure?

A) In nonhierarchical clustering, objects can be removed from an early-formed cluster and reassigned to another cluster formed later in the sequence.
B) In nonhierarchical clustering, objects cannot be removed from an early-formed cluster and reassigned to another cluster formed later in the sequence.
C) Results are more robust with nonhierarchical clustering compared to hierarchical clustering.
D) Nonhierarchical clustering is much quicker and requires less computing capability than hierarchical clustering.
E) Nonhierarchical clustering provides a simpler solution than does hierarchical clustering.

Question 37

Harriet is examining the demographic and psychographic characteristics of the individuals in each cluster in an attempt to explain why the individuals were grouped in the manner they were. Harriet is _________ clusters.

A) counting
B) profiling
C) identifying
D) scaling
E) naming

Question 38

_____ is a statistical tool that develops linear combinations of metric independent variables to predict group membership as defined by the nonmetric dependent variable.

A) Regression analysis
B) Discriminant analysis
C) Cluster analysis
D) Factor analysis
E) ANOVA

Question 39

Jose wants to predict employees likely to leave versus those who are not likely leave the company based on the number of years an individual has worked for the company and their level of satisfaction with their job. The most appropriate type of analysis for Jose to use would be _____.

A) regression analysis
B) ANOVA
C) cluster analysis
D) discriminant analysis
E) factor analysis

Question 40

Estimates of the predictive power of each independent variable in discriminant analysis are called _____.

A) z-scores
B) k-means
C) hits
D) centroids
E) discriminant weights

Question 41

The overall predictive accuracy of an estimated discriminant function is called the _____.

A) hit ratio
B) Z-score
C) discriminant weight
D) k-mean
E) centroid

Question 42

Which of the following is a guideline to answer the question of what the classification accuracy of the discriminant function should be relative to chance?

A) the classification accuracy should be at least as much as that by chance
B) the classification accuracy should be at least 25% larger than by chance
C) the classification accuracy should be at least 35% larger than by chance
D) the classification accuracy should be at least 50% larger than by chance
E) the classification accuracy should be at least 60% larger than by chance

Question 43

Factor analysis is a multivariate statistical technique that can summarize the information from a large number of variables into a much smaller number of variables or factors.

Question 44

Common factor analysis uses all the variance in the data set.

Question 45

An oblique solution means that the factors are rotated so that they are independent of one another and the correlation between them is 0.

Question 46

Latent root criterion is the most widely used method to decide the number of factors.

Question 47

Communality tells a researcher how much of the variance in a particular variable is accounted for by the factor solution.

Question 48

Cross loading is a highly desirable situation which means that there are relatively higher loadings on more than one factor.

Question 49

Cluster analysis enables researchers to combine objects into groups so that objects in each of the groups are similar to one another and different from objects in all other groups.

Question 50

Discriminant analysis is a statistical procedure that can be used to predict which group an individual is likely to belong to using two or more metric independent variables.

Question 51

The centroid is the mean of all the respondent Z scores in a particular group.

Question 52

PLS-SEM is appropriate when the theoretical model involves both reflective and formative latent variables.

Question 53

PLS-SEM cannot be used when there are missing values.

Question 54

PLS-SEM is advantageous when the main research interest is in prediction rather than theory testing.

Question 55

In PLS-SEM, the outer model is concerned with the estimation, direction and strength of the path linkages between constructs.

Question 56

In PLS-SEM, the inner model is concerned with the estimation of the loadings and the reliability and validity of the latent constructs.

Question 57

In PLS-SEM, the blindfolding procedure examines the predictive relevance of the model.

Question 58

What is factor analysis and when would the business researcher use it?

Question 59

What is the difference between dependence and interdependence techniques? Give two examples of multivariate statistical techniques for each.

Question 60

Compare and contrast cluster analysis and factor analysis.

Question 61

Compare and contrast discriminant analysis and regression analysis.

Question 62

A researcher is trying to determine factors that predict whether or not an employee is likely to stay with a company for more than five years or leave before five years, which is a nonmetric variable. She's considering factors such as attitude toward coworkers, attitude toward management, salary, and commitment toward the company, which are all measured using interval or ratio scales. Explain the appropriate statistic technique and the steps the researcher must follow when using it. What will the results tell the researcher?

Question 63

Name and describe the three types of variance in factor analysis.

Question 64

Name and describe the two basic factor analysis models, and explain when a researcher would use one over the other.

Question 65

How does the researcher decide how many factors to retain?

Question 66

Explain the three phases of cluster analysis.

Question 67

When would a researcher choose to partial least squares structural equation modeling (PLS-SEM)?