Factor Analysis (Optional Session)

Factor Analysis
1
What is Factor Analysis
 Data Reduction Technique
 A factor is a weighted sum of the variables
 The goal is to summarize the information in a larger
number of correlated variables into a smaller number of
factors that are not correlated with each other.

 In contrast to Regression, there is no dependent

variable. We just look at the correlations between
variables to summarize.

Factor Analysis
2
Graphical Intuition: Factor Analysis works
when data are correlated

C B

A D

Factor Analysis
3
 Graphical Intuition: Factor Analysis will not work
when variables are uncorrelated

Figure 2

1.0

0.9

0.8

0.7

0.6

0.5
Y

0.4

0.3

0.2

0.1

0.0

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1.0
X
Factor Analysis
4
When to do Factor Analysis in
business research?

 Applications
 Eliminating Multicollinearity problems in
Regression
 Measuring managerially useful constructs
 Intelligence, Leadership Skills, Customer
satisfaction
 Usefulin constructing perceptual maps of
products that are useful in positioning studies

Factor Analysis
5
 Perceptual Map… Example
Perceptual Map for Cars

1.5

Dodge Neon
VW Golf 1

Economy 0.5
Camry Taurus
0
-1.5 -1 -0.5 0 0.5 1 1.5 2
-0.5

-1
Lexus ES 300
-1.5 BMW325
Fashion

Factor Analysis
6
Applying Factor Analysis: Evaluating
MBA Applications

 Suppose school believes success of

future managers depends on
 Intelligence

 Teamwork and Leadership skills

Factor Analysis
7
Applying Factor Analysis:
Evaluating MBA Applications
 Variables available
 GPA
 GMAT score Which variables
 Scholarships, fellowships won
do you believe
 Evidence of Communications skills
 Prior Job Experience
correlate with
 Organizational Experience intelligence and
 Other extra curricular achievements teamwork and
leadership skills?

Factor Analysis
8
 Data…

Appli- GPA GMAT Scholar Commun Job Ex Org. Extracur

cant ship ication skills ricular
1 3.7 680 3.5 4.4 4 3 2
2
3

20

Factor Analysis
9
 Quick and dirty sense of the data: Looking at
the correlation matrix

Attribute GPA GMAT Fellowship Comm Job Ex Org Ex Extra

Curr
GPA 1.00 0.97 0.96 0.43 0.05 -0.05 -0.12
GMAT 0.97 1.00 0.99 0.55 0.27 0.16 0.12
Fellowsh 0.96 0.99 1.00 0.47 0.19 0.07 0.05
Comm 0.43 0.55 0.47 1.00 0.82 0.79 0.69
Job Ex 0.05 0.27 0.19 0.82 1.00 0.99 0.98
Org Ex -0.05 0.16 0.07 0.79 0.99 1.00 0.97
Extra Cur -0.12 0.12 0.05 0.69 0.98 0.97 1.00

Even if data is not as neatly correlated as

here… Factor analysis will be helpful

Factor Analysis
10
First Step: Do Principal Component Analysis
(PCA) to select # of factors

 PCA uses the correlation matrix of the data and

constructs factors
 Factors
 If there are n variables we will have n factors

 First factor will explain most variance, second next, and

so on…
 Variance Explained by Factors
 With standardized variables each variable has a variance

of 1, so the total variance in n variables is n

 Each factor will have an associated eigen-value which is

the amount of variance explained by that factor

Factor Analysis
11
 SPSS Output of PCA: Eigen Analysis

T otal Varian ce Explained

Initial Eigenvalues Extraction Sum s of Squared Loadings

Com ponent T otal % of Variance Cum ulative % T otal % of Variance Cum ulative %
1 3.744 53.480 53.480 3.744 53.480 53.480
2 2.268 32.398 85.878 2.268 32.398 85.878
3 .425 6.069 91.948
4 .288 4.113 96.060
5 .140 1.994 98.054
6 .098 1.406 99.460
7 .038 .540 100.000
Extraction M ethod: Principal Com ponent Analysis.

85.9% of variance in 7 variables

explained by just 2 factors
Factor Analysis
12
SPSS Output of PCA: Scree Plot

Factor Analysis
13
Second Step: Do Factor Analysis with
number of factors selected from Step 1

 First interpret resulting factors

 Use factor loadings to interpret factors
 If it is not interpretable use rotation options
until we get something that can be interpreted
 Look at factor equations and factor scores
 Score plots will be useful

Factor Analysis
14
Why not Unrotated Factor Loadings?
Variable’s correlation with the factors

Unrotated Factor Loadings and Communalities

Component Matrixa

Component
1 2
gmat .891 -.388
gpa .766 -.586
fellow .777 -.552
comm .883 .052
jobex .683 .662
organze .518 .730
extra .493 .705
Extraction Method: Principal Component Analysis.
a. 2 components extracted.
Factor Analysis
15
Interpreting Factors: Looking at
Loading Plot without Rotations
Loading Plot of GMAT-Extra without Rotations

Factor Analysis
16
 Rotated Factor Loadings: Variable’s
correlation with the factors
Rotated Factor Loadings and Communalities
Varimax Rotation a
Rotated Component Matrix

Component
1 2
gmat .954 .186
gpa .963 -.048
fellow .953 -.014
comm .698 .543
jobex .187 .933
organze .013 .895
extra .007 .860
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.
a. Rotation converged in 3 iterations.

Factor Analysis
17
Interpreting Factors: Looking at
Loading Plot with Rotation
Loading Plot of GMAT-Extra with Rotations

Factor Analysis
18
Naming Factors

 Apriori, theory based selection of

variables
 Should be easy to name factors
 Otherwise use managerial intuition

Factor Analysis
19
 How did applicants score on Intelligence and
Leadership Factors
Component Score Coefficient Matrix

Component
1 2
gmat .293 -.006
gpa .315 -.097
fellow .309 -.083
comm .181 .153
jobex -.015 .344
organze -.068 .343
extra -.068 .331
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.
Component Scores.

Intelligence=0.293 GMAT + 0.315 GPA + 0.309 Fellowships +

0.181 Communications - 0.015 Job Ex
- 0.068 Organizational Skills
- 0.068 ExtraCurricular
Leadership= - 0.006 GMAT - 0.097 GPA - 0.083 Fellowships
+ 0.153 Communications + 0.344 Job Ex
+ 0.343 Organizational Skills
+ 0.331 ExtraCurricular
Factor Analysis
20
Which Applicants to select for Haas: The
Score Plot

Too risky
1
Too Risky
Successful Applicants
Successful
applicants
0
F2Score

No Good
Sure
-1

rejects Book worms

Bookworms
-2

-2 -1 0 1 2
F1Score

Factor Analysis
21
 Step 1: Choosing number of factors to
extract from data
 Do Factor Analysis
 In SPSS select Analyze>Data Reduction>Factor…

 Select “Extraction”, select “Principle Component

Analysis”
 Select the variables you want to factor analyze in Variables
box
 Select “Correlation” as the data that will be analyzed; this will
mean that the data will be standardized and therefore each
variable will have equal effect.
 Ask for Scree Plot (using Graphs button) which graphs the
amount of variance explained by each factor

Factor Analysis
22
 Step 2: Performing Factor Analysis with #
of factors from Step 1

 Do Factor Analysis
 Number of Factors to extract should be from Step
1
 Try “None” rotation for a start (else try Varimax or
others if it doesn’t work)
 In Graphs: select loading plot and score plot

 In Storage: in the scores box store the factor

scores by selecting 2 variables

Factor Analysis
23