Ch apter Ni nete en

Factor Analysis

© 2007 Prentice Hall 19-1

 Cha pter Ou tl ine

1) Overview

2) Basic Concept

3) Factor Analysis Model

4) Statistics Associated with Factor Analysis

© 2007 Prentice Hall 19-2

 Cha pter Ou tl ine
5) Conducting Factor Analysis
i. Problem Formulation
ii. Construction of the Correlation Matrix
iii. Method of Factor Analysis
iv. Number of of Factors
v. Rotation of Factors
vi. Interpretation of Factors
vii. Factor Scores
viii. Selection of Surrogate Variables

© 2007 Prentice Hall

ix. Model Fit 19-3
 Cha pter Ou tl ine

6) Applications of Common Factor Analysis

7) Summary

© 2007 Prentice Hall 19-4

 Fac to r Ana lys is
 Fac tor anal ysi s is a general name denoting a class of
procedures primarily used for data reduction and summarization.

 Factor analysis is an int erd epe nden ce te chnique in that an

entire set of interdependent relationships is examined without
making the distinction between dependent and independent
variables.
 Factor analysis is used in the following circumstances:
 To identify underlying dimensions, or fac tors, that explain
the correlations among a set of variables.
 To identify a new, smaller, set of uncorrelated variables to
replace the original set of correlated variables in subsequent
multivariate analysis (regression or discriminant analysis).
 To identify a smaller set of salient variables from a larger set
© 2007 Prentice Hall 19-5
for use in subsequent multivariate analysis.
 Fac to rs Under lying S el ecte d
Psy chogra phi cs a nd L ifestyl es
Fi g. 19.1

Factor 2

Football Baseball

Evening at home
Factor 1
Go to a party
Home is best place

Plays
Movies
© 2007 Prentice Hall 19-6
 Fac to r Ana lys is Mo de l
Mathematically, each variable is expressed as a linear combination
of underlying factors. The covariation among the variables is
described in terms of a small number of common factors plus a
unique factor for each variable. If the variables are standardized,
the factor model may be represented as:

Xi = Ai 1 F1 + Ai 2 F2 + Ai 3 F3 + . . . + Aim Fm + Vi Ui

where

Xi = i th standardized variable
Aij = standardized multiple regression coefficient of variable
i on common factor j
F = common factor
Vi = standardized regression coefficient of variable i on
unique factor i
Ui = the unique factor for variable i
m Hall =
© 2007 Prentice number of common factors 19-7
 Fac to r An al ysi s Mo de l
The unique factors are uncorrelated with each other and
with the common factors. The common factors
themselves can be expressed as linear combinations of
the observed variables.

Fi = Wi1X1 + Wi2X2 + Wi3X3 + . . . + WikXk

Where:

Fi = estimate of i th factor
Wi = weight or factor score coefficient
© 2007 Prentice Hall
k = number of variables 19-8
 Fac to r Ana lys is Mo de l

 It is possible to select weights or factor score coefficients

so that the first factor explains the largest portion of the
total variance.

 Then a second set of weights can be selected, so that

the second factor accounts for most of the residual
variance, subject to being uncorrelated with the first
factor.

 This same principle could be applied to selecting

additional weights for the additional factors.
© 2007 Prentice Hall 19-9
 St at is tic s Asso cia ted with Facto r
Analys is

 Bar tle tt's test of sp her icity. Bartlett's test of

sphericity is a test statistic used to examine the hypothesis
that the variables are uncorrelated in the population. In
other words, the population correlation matrix is an
identity matrix; each variable correlates perfectly with
itself (r = 1) but has no correlation with the other
variables (r = 0).

 Co rrel ati on mat rix. A correlation matrix is a lower

triangle matrix showing the simple correlations, r,
between all possible pairs of variables included in the
analysis. The diagonal elements, which are all 1, are
usually
© 2007 Prentice Hall omitted. 19-10
 St at is tic s Ass ocia ted with Facto r
Analys is
 Co mmun al ity . Communality is the amount of variance a
variable shares with all the other variables being
considered. This is also the proportion of variance
explained by the common factors.
 Eig en val ue . The eigenvalue represents the total variance
explained by each factor.
 Fa ctor load ing s. Factor loadings are simple correlations
between the variables and the factors.
 Fa ctor load ing pl ot. A factor loading plot is a plot of the
original variables using the factor loadings as coordinates.
 Fa ctor mat rix. A factor matrix contains the factor
loadings of all the variables on all the factors extracted.
© 2007 Prentice Hall 19-11
 St at is tic s Ass ocia ted with Facto r
Analys is
 Fac tor sc ores. Factor scores are composite scores estimated
for each respondent on the derived factors.
 Kais er- Mey er- Olkin (K MO) me asure of samp lin g
ad equ ac y. The Kaiser-Meyer-Olkin (KMO) measure of sampling
adequacy is an index used to examine the appropriateness of
factor analysis. High values (between 0.5 and 1.0) indicate
factor analysis is appropriate. Values below 0.5 imply that factor
analysis may not be appropriate.
 Pe rc ent ag e of v ar ian ce. The percentage of the total variance
attributed to each factor.
 Re si dual s are the differences between the observed
correlations, as given in the input correlation matrix, and the
reproduced correlations, as estimated from the factor matrix.
 Scre e plot. A scree plot is a plot of the Eigenvalues against the
number of factors in order of extraction.
© 2007 Prentice Hall 19-12
 Co nduc ti ng Fa ctor An alysi s
Table 19.1 RESPONDENT
NUMBER V1 V2 V3 V4 V5 V6
1 7.00 3.00 6.00 4.00 2.00 4.00
2 1.00 3.00 2.00 4.00 5.00 4.00
3 6.00 2.00 7.00 4.00 1.00 3.00
4 4.00 5.00 4.00 6.00 2.00 5.00
5 1.00 2.00 2.00 3.00 6.00 2.00
6 6.00 3.00 6.00 4.00 2.00 4.00
7 5.00 3.00 6.00 3.00 4.00 3.00
8 6.00 4.00 7.00 4.00 1.00 4.00
9 3.00 4.00 2.00 3.00 6.00 3.00
10 2.00 6.00 2.00 6.00 7.00 6.00
11 6.00 4.00 7.00 3.00 2.00 3.00
12 2.00 3.00 1.00 4.00 5.00 4.00
13 7.00 2.00 6.00 4.00 1.00 3.00
14 4.00 6.00 4.00 5.00 3.00 6.00
15 1.00 3.00 2.00 2.00 6.00 4.00
16 6.00 4.00 6.00 3.00 3.00 4.00
17 5.00 3.00 6.00 3.00 3.00 4.00
18 7.00 3.00 7.00 4.00 1.00 4.00
19 2.00 4.00 3.00 3.00 6.00 3.00
20 3.00 5.00 3.00 6.00 4.00 6.00
21 1.00 3.00 2.00 3.00 5.00 3.00
22 5.00 4.00 5.00 4.00 2.00 4.00
23 2.00 2.00 1.00 5.00 4.00 4.00
24 4.00 6.00 4.00 6.00 4.00 7.00
25 6.00 5.00 4.00 2.00 1.00 4.00
26 3.00 5.00 4.00 6.00 4.00 7.00
27 4.00 4.00 7.00 2.00 2.00 5.00
28 3.00 7.00 2.00 6.00 4.00 3.00
29 4.00 6.00 3.00 7.00 2.00 7.00
© 2007 Prentice Hall 30 2.00 3.00 2.00 4.00 7.00 2.00 19-13
 Co nduc ti ng Fa ctor An alysi s
Fig. 19.2
Problem formulation

Construction of the Correlation Matrix

Method of Factor Analysis

Determination of Number of Factors

Rotation of Factors

Interpretation of Factors

Calculation of Selection of
Factor Scores Surrogate Variables

© 2007 Prentice Hall

Determination of Model Fit 19-14
 Conduc ting Fac tor
Analys is
For mulat e the Pro blem
 The objectives of factor analysis should be identified.

 The variables to be included in the factor analysis should

be specified based on past research, theory, and
judgment of the researcher. It is important that the
variables be appropriately measured on an interval or
ratio scale.

 An appropriate sample size should be used. As a rough

guideline, there should be at least four or five times as
many observations (sample size) as there are variables.
© 2007 Prentice Hall 19-15
 Co rre latio n Matr ix

Table 19.2

© 2007 Prentice Hall 19-16

 Conduc ti ng Fa ctor Ana lys is
Cons truc t t he Co rrela tio n
Ma trix
 The analytical process is based on a matrix of correlations
between the variables.
 Bartlett's test of sphericity can be used to test the null
hypothesis that the variables are uncorrelated in the
population: in other words, the population correlation
matrix is an identity matrix. If this hypothesis cannot be
rejected, then the appropriateness of factor analysis should
be questioned.
 Another useful statistic is the Kaiser-Meyer-Olkin (KMO)
measure of sampling adequacy. Small values of the KMO
statistic indicate that the correlations between pairs of
variables cannot be explained by other variables and that
factor analysis may not be appropriate.
© 2007 Prentice Hall 19-17
 Conduc ting Fac tor Analys is
Determ ine the Metho d of Fact or
Analys is
 In princ ipal c om pone nts analysi s, the total variance in the
data is considered. The diagonal of the correlation matrix
consists of unities, and full variance is brought into the factor
matrix. Principal components analysis is recommended when the
primary concern is to determine the minimum number of factors
that will account for maximum variance in the data for use in
subsequent multivariate analysis. The factors are called principal
components.

 In commo n fac tor anal ysis , the factors are estimated based
only on the common variance. Communalities are inserted in the
diagonal of the correlation matrix. This method is appropriate
when the primary concern is to identify the underlying
dimensions and the common variance is of interest. This method
is also known as principal axis factoring.
© 2007 Prentice Hall 19-18
 Resul ts of P ri ncipal
Com po nents Analysi s
Table 19.3
Com m una lities
Variabl es I niti al Ex tracti on
V1 1.00 0 0.926
V2 1.00 0 0.723
V3 1.00 0 0.894
V4 1.00 0 0.739
V5 1.00 0 0.878
V6 1.00 0 0.790

I nitial Eig en values

Factor Eigen va lue % of varianc e Cum ulat. %
1 2.731 45.52 0 4 5.520
2 2.218 36.96 9 8 2.488
3 0.442 7.360 8 9.848
4 0.341 5.688 9 5.536
5 0.183 3.044 9 8.580
6 0.085 1.420 10 0.0 00

© 2007 Prentice Hall 19-19

 Resul ts of P ri ncipal
Com po nents Anal ysi s
Table 19.3, cont.

© 2007 Prentice Hall 19-20

 Resul ts of P ri ncipal Co mponen ts
Analys is
Table 19.3, cont.

© 2007 Prentice Hall 19-21

 Resul ts of P ri ncipal
Com po nents Analysi s
Table 19.3, cont.

The lower-left triangle contains the reproduced

correlation matrix; the diagonal, the communalities;
the upper-right triangle, the residuals between the
observed correlations and the reproduced
correlations.

© 2007 Prentice Hall 19-22

 Co nduc ti ng Fa ctor An alysi s
De ter mine t he Number o f
Fac to rs
 A Pr iori De terminat ion . Sometimes, because of prior
knowledge, the researcher knows how many factors to expect
and thus can specify the number of factors to be extracted
beforehand.

 De termination B as ed o n Eig env al ues. In this approach,

only factors with Eigenvalues greater than 1.0 are retained. An
Eigenvalue represents the amount of variance associated with
the factor. Hence, only factors with a variance greater than 1.0
are included. Factors with variance less than 1.0 are no better
than a single variable, since, due to standardization, each
variable has a variance of 1.0. If the number of variables is less
than 20, this approach will result in a conservative number of
factors.
© 2007 Prentice Hall 19-23
 Co nduc ti ng Fa ctor An alysi s
De termi ne th e Number o f
Fac to rs
 Det er mi na tio n Based on Scr ee Pl ot. A scree plot
is a plot of the Eigenvalues against the number of factors
in order of extraction. Experimental evidence indicates
that the point at which the scree begins denotes the true
number of factors. Generally, the number of factors
determined by a scree plot will be one or a few more than
that determined by the Eigenvalue criterion.

 Det er mi na tio n Based on Pe rce nt age of Va ri an ce.

In this approach the number of factors extracted is
determined so that the cumulative percentage of variance
extracted by the factors reaches a satisfactory level. It is
recommended that the factors extracted should account
for at least 60% of the variance.
© 2007 Prentice Hall 19-24
 Sc re e Plo t
Fig. 19.3
3.0

2.5

2.0
Eigenvalue

1.5

1.0

0.5

0.0
1 2 3 4 5 6
© 2007 Prentice Hall
Component Number 19-25
 Co nduc ti ng Fa ctor An alysi s
De termi ne th e Number o f
Fac to rs
 Det er mi na tio n Based on Split -Ha lf Rel ia bi lit y.
The sample is split in half and factor analysis is
performed on each half. Only factors with high
correspondence of factor loadings across the two
subsamples are retained.

 Det er mi na tio n Based on Signi fican ce Test s. It is

possible to determine the statistical significance of the
separate Eigenvalues and retain only those factors that
are statistically significant. A drawback is that with large
samples (size greater than 200), many factors are likely
to be statistically significant, although from a practical
viewpoint many of these account for only a small
proportion of the total variance.
© 2007 Prentice Hall 19-26
 Co nduc ti ng F act or An alysis
Ro ta te Fa ctors
 Although the initial or unrotated factor matrix indicates the
relationship between the factors and individual variables, it
seldom results in factors that can be interpreted, because
the factors are correlated with many variables. Therefore,
through rotation the factor matrix is transformed into a
simpler one that is easier to interpret.
 In rotating the factors, we would like each factor to have
nonzero, or significant, loadings or coefficients for only
some of the variables. Likewise, we would like each
variable to have nonzero or significant loadings with only a
few factors, if possible with only one.
 The rotation is called orthogonal r ota tio n if the axes
are maintained at right angles.
© 2007 Prentice Hall 19-27
 Co nduc ti ng F act or An alysis
Ro ta te Fa ctors
 The most commonly used method for rotation is the
va rima x pr oce du re. This is an orthogonal method of
rotation that minimizes the number of variables with high
loadings on a factor, thereby enhancing the
interpretability of the factors. Orthogonal rotation
results in factors that are uncorrelated.
 The rotation is called obliq ue rotat ion when the axes
are not maintained at right angles, and the factors are
correlated. Sometimes, allowing for correlations among
factors can simplify the factor pattern matrix. Oblique
rotation should be used when factors in the population
are likely to be strongly correlated.
© 2007 Prentice Hall 19-28
 Fac tor Mat ri x Befo re and Aft er
Ro tat ion
Fig. 19.4

Fac to rs Fac tor s

Variables 1 2 Variables 1 2
1 X 1 X
2 X X 2 X
3 X 3 X
4 X X 4 X
5 X X 5 X
6 X 6 X
(a) (b)
High Loadings High Loadings
Before Rotation After Rotation
© 2007 Prentice Hall 19-29
 Co nduc ti ng F actor An alysis
Inte rpre t F ac to rs

 A factor can then be interpreted in terms of the

variables that load high on it.
 Another useful aid in interpretation is to plot the
variables, using the factor loadings as coordinates.
Variables at the end of an axis are those that
have high loadings on only that factor, and hence
describe the factor.

© 2007 Prentice Hall 19-30

 Fac to r Lo adi ng Pl ot
Rota ted Componen t Matri x
Fig. 19.5 Compon ent 2
Comp onen t Plot Com pon ent
in Vari abl e 1
Comp
Rot onen
ated t 1ce
Spa 2
1.0 ∗∗ ∗ V6
V4
V2 V1 0.9 62
-2.6 6E-02
0.5
V2 -5.7 2E-02 0.8 48

0.0 ∗V1
V3 0.9 34
∗ V5 V3 ∗ -0.1 46
-0.5 V4 -9.8 3E-02 0.8 54
V5 -0.9 33
-1.0 -8.4 0E-02
1.0 0.5 0.0 -0.5 -1.0 V6 8.3 37 E-02 0.8 85
© 2007 Prentice Hall 19-31
 Co nduc ti ng F act or
An alysi s
Ca lcu late F acto r S cores

The facto r sc ores for the ith factor may be

estimated
as follows:

Fi = Wi1 X1 + Wi2 X2 + Wi3 X3 + . . . + Wik Xk

© 2007 Prentice Hall 19-32

 Co nduc ti ng F act or
An alysi s
Sel ect S ur rogate Va ri abl es
 By examining the factor matrix, one could select for
each factor the variable with the highest loading on
that factor. That variable could then be used as a
surrogate variable for the associated factor.
 However, the choice is not as easy if two or more
variables have similarly high loadings. In such a
case, the choice between these variables should be
based on theoretical and measurement
considerations.
© 2007 Prentice Hall 19-33
 Co nduc ti ng F actor
An al ysi s
De ter mine the Mo del Fit

 The correlations between the variables can be

deduced or reproduced from the estimated
correlations between the variables and the factors.

 The differences between the observed correlations

(as given in the input correlation matrix) and the
reproduced correlations (as estimated from the
factor matrix) can be examined to determine
model fit. These differences are called residuals.

© 2007 Prentice Hall 19-34

 Resu lts of Com mo n
Fac to r An al ysi s
Table 19.4

Ba rl ett tes t of sph eric it y

• Approx. Chi-Square = 111.314
• df = 15
• Significance = 0.00000
• Kaiser-Meyer-Olkin measure of
sampling adequacy = 0.660

© 2007 Prentice Hall 19-35

 Resu lts of Com mo n
Fac to r An al ysi s
Table 19.4, cont.

© 2007 Prentice Hall 19-36

 Res ul ts o f Com mo n
Fac to r An al ysi s
Table 19.4, cont.

© 2007 Prentice Hall 19-37

 Resu lts of Co mm on
Fac to r Ana lys is
Table 19.4, cont.

© 2007 Prentice Hall 19-38

 SPSS Wi ndow s
To select this procedure using SPSS for Windows
click:

An aly ze> Da ta Red ucti on>Fact or …

© 2007 Prentice Hall 19-39

 SPSS Wi ndo ws: Princ ipal
Com po ne nts
1. Select ANALYZE from the SPSS menu bar.
2. Click DATA REDUCTION and then FACTOR.
3. Move “Prevents Cavities [v1],” “Shiny Teeth [v2],” “Strengthen Gums
[v3],” “Freshens Breath [v4],” “Tooth Decay Unimportant [v5],” and
“Attractive Teeth [v6].” in to the VARIABLES box..
4. Click on DESCRIPTIVES. In the pop-up window, in the STATISTICS box
check INITIAL SOLUTION. In the CORRELATION MATRIX box check KMO
AND BARTLETT’S TEST OF SPHERICITY and also check REPRODUCED.
Click CONTINUE.
5. Click on EXTRACTION. In the pop-up window, for METHOD select
PRINCIPAL COMPONENTS (default). In the ANALYZE box, check
CORRELATION MATRIX. In the EXTRACT box, check EIGEN VALUE OVER
1(default). In the DISPLAY box check UNROTATED FACTOR SOLUTION.
Click CONTINUE.
6. Click on ROTATION. In the METHOD box check VARIMAX. In the
DISPLAY box check ROTATED SOLUTION. Click CONTINUE.
7. Click on SCORES. In the pop-up window, check DISPLAY FACTOR SCORE
COEFFICIENT MATRIX. Click CONTINUE.
© 2007 Prentice Hall 19-40
8. Click OK.