Deck 9: Factor Analysis

Question 1

Factor analysis is used to describe the underlying structure that explains a set of variables.

Answer 1

True

Question 2

The underlying hypothetical (unobservable) variables in factor analysis are called factors.

Answer 2

True

Question 3

The main set of results obtained from a factor analysis consists of factor loadings.

Answer 3

True

Question 4

An index provided inn the results of a factor analysis is the list of communalities for each variable.

Question 5

The process by which the factors are determined from a larger set of variables is called extraction.

Question 6

In principal components analysis, only unique variability is analyzed for each observed variable.

Question 7

In factor analysis, unique, shared, and error variability is analyzed for each observed variable.

Question 8

Factor analysis analyzes variance.

Question 9

Principle components analysis analyzes covariance.

Question 10

Principal components analysis is usually the preferred method of factor extraction, especially when the focus of an analysis searching for an underlying structure is explanatory.

Question 11

Kaiser's rule states that only those components in principal components analysis whose eigenvalues are greater than 1 should be retained.

Question 12

An eigenvalue is defined as the amount of total variance explained by each factor, with the total amount of variability in the analysis equal to the number of original variables in the analysis.

Question 13

A scree plot is a graph of the magnitude of each eigenvalue (vertical axis) plotted against its ordinal numbers (horizontal axis).

Question 14

A general rule of thumb is to retain the factors that account for at least 70% of the total variability.

Question 15

A final criterion for retaining components is the assessment of model fit.

Question 16

Rotation is a process by which a factor solution is made more interpretable by altering the underlying mathematical structure.

Question 17

Orthogonal rotation is a rotation of factors that results in factors being correlated with each other.

Question 18

Oblique rotation results in factors being uncorrelated with each other.

Question 19

Varimax is the most commonly used oblique rotation procedure.

Question 20

A factor correlation matrix is produced from an orthogonal rotation.

Question 21

Interpretation of components or factors involves much subjective decision making on the part of the researcher.

Question 22

When interpreting or naming components, one should pay particular attention to the size and direction of each loading.

Question 23

A bipolar factor refers to a component that contains both high positive and high negative loadings.

Question 24

Principal components analysis may be used as a variable reducing scheme for further analysis.

Question 25

Factor scores are estimates of the scores participants would have received on each of the factors had they been measured directly.

Question 26

There are two basic types of factor analytic procedures, based on their overall intended function. They include:

A) Exploratory factor analysis.
B) Explanatory factor analysis.
C) Confirmatory factor analysis.
D) Both (a) and (b) are correct.

Question 27

In exploratory factor analysis, the goal is to:

A) Describe data by grouping together variables that are correlated.
B) Summarize data by grouping together variables that are uncorrelated.
C) Describe and summarize data by grouping together variables that are correlated.
D) Test a theory about latent processes that might occur among variables.

Question 28

Confirmatory factor analysis often used to:

A) Test a theory about underlying, unobservable processes that might occur among variables.
B) Confirm or disconfirm a theory post hoc.
C) Neither (a) nor (b).
D) Both (a) and (b).

Question 29

It is recommended that the following two assumptions be evaluated and any necessary transformations be made to ensure the quality of data and improve the quality of the resulting factor or component solution:

A) All variables, as well as all linear combinations of variables, must be normally distributed.
B) The relationships among all variables must be linear.
C) The relationships among all pairs of variables must be linear.
D) Both (a) and (c) are correct.

Question 30

The underlying, mathematical objective in principal components analysis is to obtain:

A) Correlated linear combinations of the original variables that account for as much of the total variance in the original variables as possible.
B) Uncorrelated linear combinations of the original variables that account for some of the total variance in the original variables.
C) Uncorrelated linear combinations of the original variables that account for as much of the total variance in the original variables as possible.
D) Uncorrelated combinations of the original variables that account for as much of the total variance in the original variables as possible.