chapter 17

Question 1

what is the difference between latent and observed variables

Answer 1

observed are- observed and latent represent an underlying construct that is not directly measured and inferred by the observed variables

Question 2

Factor analysis attempts to achieve parsimony by explaining the __________ amount of ____ ________ in a correlation matrix using the _______ _____ of explanatory constructs.

Answer 2

Factor analysis attempts to achieve parsimony by explaining the MAXIMUM amount of COMMON VARIANCE in a correlation matrix using the SMALLEST NUMBER of explanatory constructs.

Question 3

Explanatory constructs in FA are called

Answer 3

factors

Question 4

PCA tries to explain the maximum amount of..

Answer 4

total variance in a correlation matrix

Question 5

How does PCA explain total variance in a correlation matrix

Answer 5

by transforming variables into linear components

Question 6

Exploratory factor analysis will

Answer 6

describe and summarize

Question 7

confirmatory factor analysis will

Answer 7

test hypothesis and structure

Question 8

How many steps in FA and PCA

Answer 8

7

Question 9

Steps in PCA and FA:

1. Select and measure___ ___ ____
2. Make a ________ ______
3. Extract a set of _______ from the correlation matrix
4. Determine the number of________
5. ______ the factors
6. _______ _____ _______
7. Verify the ______ _______

Answer 9

Steps in PCA and FA:

1. Select and measure A SET OF VARIABLES
2. Make a CORRELATION MATRIX
3. Extract a set of FACTORS from the correlation matrix
4. Determine the number of FACTORS
5. ROTATE the factors
6. INTERPRET THE RESULTS
7. Verify the FACTOR STRUCTURE

Question 10

Why do we rotate the factors

Answer 10

to increase interpretability

Question 11

IN FA AND PA we seek to ________ the R-matrix into a smaller set of ______ dimensions.

Answer 11

reduce the r-matrix into a smaller set of uncorrelated dimensions

Question 12

True or false: The assumption f FA is that algebraic factors in a factor matrix represent real - world dimensions

Answer 12

True

Question 13

True or false: PA CANNOT be used to solve the problem of multicollinearity

Answer 13

FALSE_ FA can be used to solve multicollinearity problems

Question 14

Both PA and FA look for variables that correlate highly with one another but…

Answer 14

not with anything else.

Question 15

The factor loading can be thought of as the…

Answer 15

Pearson correlation between a factor and a variable

Question 16

So if we square a factor loading we get the

Answer 16

substantive importance of a particular variable

Question 17

In a matrix columns represent each_________ and rows represent

Answer 17

factor and rows represent each variable on each factor.

Question 18

Weighted average simply puts a persons scores into…

What will influence the resulting scores?

Answer 18

• the equation and gives you a score on those particular factors.
• scales of measurement.

Question 19

If scales of measurement are different what does it mean

Answer 19

we cannot compare factor scores with different scales of measurement.

Question 20

Recommended sample size for FA=
and minimum number of responses=
minimum correlation r value=

Answer 20

• at least 300
• 5 to 10
• .3

Question 21

Factor analysis is usually performed on

Answer 21

ordinal or continuous

variables

Question 22

Bartlett Method and Anderson Rubin Method are used to calculate

Answer 22

factor scores

Question 23

Bartlett Method and Anderson Rubin Method are used to calculate factor scores instead of regression because the regression method

Answer 23

means the scores can correlate with other factors

Question 24

If you want factor scores that are uncorrelated and standardized use the

Answer 24

Anderson Rubin Method

Question 25

How can results of PCA be generalized to the population..

This is called

Answer 25

if analysis using a different sample confirms the factor structure.
cross-validation

Question 26

Kaiser recommended retaining all factors with

Answer 26

eigenvalues greater than 1

Question 27

An eigenvalue of 1 represents

Answer 27

a substantial amount of variation

Question 28

Orthogonal means

Answer 28

unrelated

Question 29

when we rotate something on an orthogonal axis we rotate but

Answer 29

keep factors independent

Question 30

Oblique allows for

Answer 30

correlation between factors

Question 31

Orthogonal rotations include

Answer 31

varimax
equinox
Quartmax

Question 32

Direct oblivion and promax rotations are

Answer 32

oblique

Question 33

As commonalities lower sample size

Answer 33

becomes of greater importance

Question 34

If you have a KMO value of 0

Answer 34

factor analysis is inappropriate and there are a large number of semi partial correlations

Question 35

KMO value below .50s are

Answer 35

no good

Question 36

Bartlett’s test tells us whether our correlation matrix is significantly different than…
or in other words:
That the correlations between variables are …

Answer 36

…an identity matrix

…significantly different than zero

Question 37

We want Bartlett’s test to be

Answer 37

significant

Question 38

proportion of common variance in a variable is called

Answer 38

the communality

Question 39

Communality of 1 =

Answer 39

all shared variance

Question 40

Communality of 0=

Answer 40

zero shared.

Question 41

The extraction column of the commonalities table can be multiplied by 100 and expressed as a percentage of…

Answer 41

shared variance or common variance