ASSIGNMENTTT

NO 3
The best solution for a set of components/factors

Multivariate Data Analysis

Exploratory Factor
Analysis
2021
Factor analysis is a technique that is used to reduce a large number of
variables into fewer numbers of factors. This technique extracts maximum
common variance from all variables and puts them into a common score.

Submitted to: Sir Dr. Qlander Hayat

Submitted by: Arooj fiaz
Roll no: 13380 (MBA 6, AFTERNOON)
NATIONAL UNIVERSITY OF MODERN
LANGUAGES, ISLAMABAD
Contents
Conducting the Factor Analysis Using SPSS.......................................................................................1
STANDARD DEVIATION:..................................................................................................................2
CORRELATION MATRIX..................................................................................................................4
KMO AND BARTLETT'S TEST.........................................................................................................8
COMMUNALITIES..............................................................................................................................9
TOTAL VARIANCE EXPLAINED...................................................................................................11
SCREE PLOT:....................................................................................................................................14
COMPONENT MATRIX...................................................................................................................15
ROTATED COMPONENT MATRIX...............................................................................................18
REPRODUCED CORELATION:......................................................................................................22
COMPONENT PLOT IN ROTATED SPACE.................................................................................23
CONCLUSION:...................................................................................................................................23
EXPLORATORY FACTOR ANALYSIS

EXPLORATORY FACTOR ANALYSIS:

Factor analysis is frequently used to develop questionnaires. The primary use of factor analysis
in the development of questionnaires is to ensure that questions designed to measure an ability or
trait are in fact related to the construct that you intend to measure.

There are four basic steps to the factor analysis process:

1. Data screening: calculate the descriptive statistics and a correlation matrix of all
variables to be used in the analysis.
2. Extract factors
3. Rotate factors to create a more understandable factor structure
4. Interpret results

Conducting the Factor Analysis Using SPSS

The first part of the analysis consists of determining the number of extracted factors.

1. Click Analyze, Click Data Reduction and Click Factor. You should see the Factor
Analysis dialogue box.
2. Holding down the control key, click the 58 SPSS variables (items a through j). The
click on > to move the items to the Variables box in the Factor Analysis dialog box.
3. At the bottom of this box, click on Descriptives. In the Descriptives dialog box make
sure that Univariate descriptives, Coefficients, Significance Levels, Determinant,
KMO, Barlett’s test of Sphericity and Reproduced are checked. Now click on
Continue.
4. Click Extraction. You will see the “Factor Analysis: Extraction” dialog box.
5. In the “Extraction” box you will notice that the default Method is Principal
Components. This is the most commonly used method for exploratory purposes.
6. Continuing in “Extraction” box, we also want to make sure that the boxes
Correlation Matrix, Unrotated Factor Solution, and Eignevalues over: 1 and
Scree plot are checked. Now click on Continue.

Rotating Factors

7. When back in the Factor Analysis dialogue box, click Rotation. You will see the
Rotation Dialog box.
8. In the Rotation dialog box make sure that you check Varimax. Varimax is a type of
orthogonal rotation method. Make sure that the Rotated Solution and Loading plots
boxes are checked. Change the maximum iterations to “30”. (Normally 25 is
sufficient but we have an unusually large dataset to work with in this example). Now
click on Continue
9. Now back in the Factor Analysis dialogue box, click Options.

Page | 3
EXPLORATORY FACTOR ANALYSIS

10. In the Factor Analysis: Options box under Missing Values select Exclude Cases
Pairwise. Now under Coefficient Display Format: Make sure that you check Sorted
by Size. Click on Continue.
11. From the Factor Analysis dialogue box click on Scores. In the Factor Analysis:
Scores dialogue box check Save as variables and under Save as Variables select
Anderson-Rubin. Next check Display Factor Score Coefficient Matrix. Click on
Continue.

In the Factor Analysis dialogue box, click Continue and OK

STANDARD DEVIATION:

Std. Analysis Missing

Mean Deviation N N
TEC 2.81 1.211 170 0
1
TEC 3.08 1.221 170 0
2
TEC 2.96 1.213 170 0
3
TEC 3.05 1.268 170 0
4
TEC 3.11 1.236 170 0
5
TEC 3.21 1.265 170 0
6
TEC 3.48 1.232 170 0
7
TEC 3.45 1.283 170 0
8
PA1 2.92 1.096 170 0
PA2 3.22 .964 170 0
PA3 3.28 .967 170 0
PA4 3.41 .989 170 0
PA5 3.32 1.118 170 0
PA7 3.08 1.066 170 0
PA8 3.16 1.111 170 0
PA9 3.10 1.155 170 0
PA10 3.14 1.114 170 0

Page | 4
EXPLORATORY FACTOR ANALYSIS

PA11 3.11 1.148 170 0

PA12 3.39 1.111 170 0
PA14 3.21 1.212 170 0
PA15 3.39 1.147 170 0
SR2 3.27 1.070 170 0
SR3 3.28 .991 170 0
SR4 3.32 1.052 170 0
SR5 3.19 1.109 170 0
SR7 3.19 1.106 170 0
SR8 3.26 1.118 170 0
SR9 3.25 1.166 170 0
SR10 3.32 1.138 170 0
SR11 3.32 1.194 170 0
SR12 3.19 1.142 170 0
FC3 2.92 1.096 170 0
FC4 3.11 1.141 170 0
FC6 3.01 1.164 170 0
FC7 3.24 1.164 170 0
CBI4 3.61 .969 170 0
CBI6 3.67 .990 170 0
CBI7 3.75 .967 170 0
CBI8 3.49 1.039 170 0
CBI9 3.54 1.094 170 0
SR6 3.29 1.148 170 0
PA16 3.61 1.132 170 0
FC5 3.05 1.220 170 0
PA6 3.25 1.115 170 0
PA13 3.39 1.122 170 0
SR1 3.05 1.153 170 0
FC8 3.04 1.158 170 0
CBI5 3.55 .967 170 0
CBI1 3.34 1.088 170 0
CBI2 3.54 .998 170 0

Page | 5
EXPLORATORY FACTOR ANALYSIS

CORRELATION MATRIX

Screenshots of correlation matrix

Page | 6
EXPLORATORY FACTOR ANALYSIS

Page | 7
EXPLORATORY FACTOR ANALYSIS

Page | 8
EXPLORATORY FACTOR ANALYSIS

OUTPUT 1:

Correlations
The first step is to examine the correlation matrix (refer to output) between variables (items) to
examine how well they relate to one another. If we find that there are variables that do not
correlate well with any other variables (or very few) then we should consider excluding these
variables before the factor analysis is conducted. We would like to see our correlation
coefficients exceed .30.
Multicollinearity (Singularity)
The opposite problem of low correlations is variable that correlate too highly. It is important to
avoid extreme multicollinearity (i.e. variable that are very highly correlated) or singularity
(variables that are perfectly correlated). As with regression, singularity causes problems in factor
analysis because it becomes impossible to determine the unique contribution of a variable that is
highly correlated with another variable.
Bottom Line for Treating Collinearity
At this early stage we look to eliminate any variables that show no relationship (do not correlate)
with any other variables or that correlate too highly with other variables (i.e. r > .90).

KMO AND BARTLETT'S TEST

KMO and Bartlett's Test

Kaiser-Meyer-Olkin Measure of Sampling .753
Adequacy.
Bartlett's Test of Approx. Chi-Square 3835.27
Sphericity 0
Df 1225
Sig. .000

OUTPUT 2:

To evaluate the issues of low correlations and Singularity refer to the Correlation Matrix in
addition to KMO (Kaiser-Meyer-Olkin) and Barlett’s Test of Sphericity sections of the SPSS
output.

1. Examine the correlations between variables

2. Notice a Determinant value is listed at the bottom of the correlation matrix. The value for
the determinant is an important test for multicollinearity or singularity. The determinant
of the correlation matrix should be greater than .00001. If the Determinant value is less
than this value, it would be important to attempt to identify pairs of variables where

Page | 9
 Communalities
Extractio
EXPLORATORY FACTOR ANALYSIS Initial n
TEC 1.000 .705
r > .8 and consider 1 eliminating them from the
analysis. TEC 1.000 .655
3. The Barlett’s test is 2 designed to determine if the
correlation matrix is an identity matrix (where all
TEC 1.000 .571
correlation coefficients are 0). A significant value (less
than .05) indicates that 3 the data do not produce an
identity matrix TEC 1.000 .625 indicating there are adequate
relationships between 4 variables to conduct the factor
analysis. Results from TEC 1.000 .536 this test also indicate that the
correlations among 5 variables overall are not so
strong suggesting multicollinearity.
TEC 1.000 .574
6
4. The KMO test is a TEC 1.000 .669 measure of whether the
distribution of values 7 based on the sample is
adequate for conducting TEC 1.000 .557 a factor analysis. This test
indicates the amount of overlap or shared variance
between pairs of 8 variables (remember we are
trying to identify items PA1 1.000 .699 that are related but yet provide
unique information to PA2 1.000 .626 the factors we are attempting to
identify). Values PA3 1.000 .627 should be greater than .5.The
value is .753 which is PA4 1.000 .565 good.
PA5 1.000 .494
COMMUNALITIES
PA7 1.000 .505
PA8 1.000 .500
PA9 1.000 .523
OUTPUT 3: PA10 1.000 .571
Refer to the output entitled PA11 1.000 .692 “Communalities”.
Communalities are estimates PA12 1.000 .572 of shared or common variance
PA14 1.000 .504
among the variables after extraction has taken place.
PA15 1.000 .555
Communalities for each SR2 1.000 .618 variable can also be interpreted
as the squared multiple SR3 1.000 .619 correlation (R2) of the variable
predicted from the combination SR4 1.000 .643 of extracted factors. The goal
SR5 1.000 .618
of factor analysis is to identify groups of variables (items in
SR7 1.000 .554
this case) that are related to one another and derive a
SR8 1.000 .631
description of the underlying SR9 1.000 .641 traits that best represent the
data structure SR10 1.000 .700
SR11 1.000 .605
TOTAL VARIANCE SR12 1.000 .685 EXPLAINED
FC3 1.000 .606
FC4 1.000 .502
FC6 1.000 .544 Page | 10
FC7 1.000 .553
CBI4 1.000 .606
CBI6 1.000 .663
EXPLORATORY FACTOR ANALYSIS

Total Variance Explained

Extraction Sums of Rotation Sums of
Initial Eigenvalues Squared Loadings Squared Loadings
Com % of Cumu % of Cumu % of Cumul
pone Tot Varia lative Tota Varia lative Tota Varia ative
nt al nce % l nce % l nce %
1 9.5 19.16 19.16 9.58 19.16 19.16 4.23 8.471 8.471
84 7 7 4 7 7 6
2 3.9 7.956 27.12 3.97 7.956 27.12 3.60 7.207 15.67
78 3 8 3 3 8
3 3.1 6.244 33.36 3.12 6.244 33.36 3.33 6.662 22.34
22 7 2 7 1 0
4 2.3 4.605 37.97 2.30 4.605 37.97 2.94 5.882 28.22
03 2 3 2 1 2
5 2.0 4.043 42.01 2.02 4.043 42.01 2.66 5.337 33.55
21 5 1 5 8 9
6 1.7 3.440 45.45 1.72 3.440 45.45 2.47 4.958 38.51
20 5 0 5 9 7
7 1.5 3.191 48.64 1.59 3.191 48.64 2.39 4.791 43.30
95 5 5 5 6 8
8 1.5 3.018 51.66 1.50 3.018 51.66 2.23 4.460 47.76
09 3 9 3 0 8
9 1.4 2.836 54.49 1.41 2.836 54.49 2.14 4.280 52.04
18 9 8 9 0 9
10 1.3 2.758 57.25 1.37 2.758 57.25 1.97 3.949 55.99
79 7 9 7 5 8
11 1.2 2.592 59.84 1.29 2.592 59.84 1.92 3.851 59.84
96 9 6 9 5 9
12 1.2 2.489 62.33
44 8
13 1.1 2.303 64.64
51 1
14 1.1 2.240 66.88
20 1
15 . 1.995 68.87
998 6
16 . 1.931 70.80
965 7

Page | 11
EXPLORATORY FACTOR ANALYSIS

17 . 1.890 72.69
945 7
18 . 1.827 74.52
913 4
19 . 1.655 76.17
827 9
20 . 1.536 77.71
768 4
21 . 1.523 79.23
762 8
22 . 1.487 80.72
743 5
23 . 1.397 82.12
699 2
24 . 1.285 83.40
642 6
25 . 1.201 84.60
600 7
26 . 1.143 85.75
572 1
27 . 1.060 86.81
530 1
28 . .986 87.79
493 7
29 . .951 88.74
475 8
30 . .885 89.63
443 3
31 . .835 90.46
418 8
32 . .808 91.27
404 6
33 . .760 92.03
380 6
34 . .729 92.76
364 5
35 . .710 93.47
355 5

Page | 12
EXPLORATORY FACTOR ANALYSIS

36 . .633 94.10
316 8
37 . .608 94.71
304 7
38 . .593 95.31
297 0
39 . .557 95.86
279 7
40 . .534 96.40
267 2
41 . .497 96.89
249 9
42 . .489 97.38
245 8
43 . .447 97.83
224 6
44 . .433 98.26
216 9
45 . .378 98.64
189 7
46 . .317 98.96
158 4
47 . .292 99.25
146 6
48 . .275 99.53
138 1
49 . .248 99.77
124 9
50 . .221 100.0
111 00

Page | 13
EXPLORATORY FACTOR ANALYSIS

Extraction Method: Principal Component Analysis.

OUTPUT 4:
The total amount of variance for a component or factor is represented as an
Eigenvalue. The eigenvalue for the first component is 9.584 and accounts for of
19.167the variability or variance of the total data structure .Components or
factors with eigenvalues of “1” or greater are considered to contribute
significantly to the data structure. SPSS by default extracts only components or
factors with eigenvalues of “1” or greater.

 In the case of factor analysis we would like to explain at least 60-

70% of the variance in the data structure by the factors that are
extracted. How much total variance in the data structure is
explained by this example (refer to your “Total Variance Explained”
table).

SCREE PLOT:

OUTPUT 5:
If there are less than 30 variables and communalities after extraction are
greater than 0.7 or if the sample size exceeds 250 and the average
communality is greater than 0.6 then retain all factors with Eigen values

Page | 14
EXPLORATORY FACTOR ANALYSIS

above 1 (Kaiser’s criterion).

If none of the above apply, a Scree Plot can be used when the sample size
is large (around 300 or more cases).

COMPONENT MATRIX

Component Matrixa
Component
1 2 3 4 5 6 7 8 9 10 11
PA .582
16
SR .572
10
SR .568
12
SR .566
7
PA .564
15
SR .558
2
SR .554
1
SR .553 .417
3
TE .543
C8
PA .543
4
SR .542
11
CB .532
I4
SR .532
6
TE .529
C3
CB .528
I8

Page | 15
EXPLORATORY FACTOR ANALYSIS

CB .524
I9
SR .521
8
PA .520
5
SR .519
5
SR .518
9
TE .510
C6
PA .507
2
SR .496
4
CB .495
I1
CB .495
I7
CB .491
I6
CB .490 -.405
I5
CB .484 -.475
I2
PA .480
3
TE .466
C4
TE .455
C5
TE .453
C7
PA .607 .424
11
FC .598
5

Page | 16
EXPLORATORY FACTOR ANALYSIS

FC .554
4
FC .516
6
PA .508
14
PA .506
9
FC .497
3
PA .470
6
PA .420
7
PA .413 .406
10
PA
8
PA .405 .512
12
PA
13
FC
8
FC
7
PA .522
1
TE .403 -.404 .421
C2
TE .423
C1
Extraction Method: Principal Component Analysis.
a. 11 components extracted.

Page | 17
EXPLORATORY FACTOR ANALYSIS

ROTATED COMPONENT MATRIX

Rotated Component Matrixa

Component
1 2 3 4 5 6 7 8 9 10 11
SR .759
10
SR .717
8
SR .689
9
SR .660
12
SR .648
7
SR .566
11
SR .522
2
TE .752
C7
TE .698
C6
TE .670
C8
TE .596
C5
TE .507
C4
CB .774
I6
CB .730
I7
CB .684
I8

Page | 18
EXPLORATORY FACTOR ANALYSIS

CB .565
I4
CB .489
I9
PA .650
6
PA .642
7
PA .642
8
PA .590
9
PA .585
14
FC .743
3
FC .647
5
FC .632
6
FC .598
4
FC .533
8
FC .490
7
SR .616
4
SR .613
5
SR .545
3
SR .504
6
PA .456
16
PA .752
1

Page | 19
EXPLORATORY FACTOR ANALYSIS

SR .586
1
PA .542
2
PA .695
3
PA .488
4
PA .467
5
PA .440
15
CB .674
I2
CB .596
I1
CB .433
I5
PA .614
11
PA .542
12
PA .508
13
PA .503
10
TE .774
C1
TE .558
C2
TE .491
C3
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.a
a. Rotation converged in 32 iterations.

OUTPUT 6:

Rotated Component (Factor) Matrix

Page | 20
 EXPLORATORY FACTOR ANALYSIS

The Rotated Component (Factor) Matrix table in SPSS provides the Factor Loadings for
each variable (in this case item) for each factor. A Factor Loading is the Pearson correlation (r)
coefficient between the original variable with a factor. For example, if we consider question 6,
we can see that it “loads on” or correlates .800 with Component 1 (Factor 1), -.010 with
Component 2 (Factor 2) and .097 with Component 3 (Factor 3) and -.072 with Component 4
(Factor4).

 A factor loading of + .50 or greater is consider practically significant with a sample

size of approximately 100. Of course the larger the sample size the smaller the loadings
need to be for practical significance.
We will continue to identify which items are highly correlated with factors and attempt to find
the patterns of items which represent factors. The next step, is to examine the items to determine
if they represent some type of attribute or trait that can be use to name the factor. In our example,
11 factors have emerged. SR,7,8,9,10,11,2 load on Factor 1. TEC 4,5,6,7,8 load on Factor 2 and
so on.

COMPONENT TRANSFORMATION MATRIX

Component Transformation Matrix

Compo
nent 1 2 3 4 5 6 7 8 9 10 11
1 .512 .437 .416 .098 .026 .320 .294 .292 .202 .064 .209
2 .154 -.281 .146 .611 .592 .014 -.096 -.105 -.065 .290 -.208
3 -.508 .283 .070 .482 -.339 -.256 .190 .118 .011 .413 .151
4 .355 .335 -.578 .264 -.154 .069 -.059 .086 -.548 -.022 -.131
5 -.099 .531 .419 -.200 .296 -.359 -.312 -.152 -.382 -.088 -.048
6 -.415 .091 -.153 .180 .351 .334 -.245 .435 .006 -.400 .347
7 -.184 .101 -.232 -.224 .467 -.116 .766 .061 -.109 -.004 -.129
8 .051 .423 -.438 -.036 .236 -.070 -.218 -.317 .613 .192 .057
9 .299 -.145 -.062 .245 .021 -.661 .116 .015 .101 -.418 .436
10 -.104 .049 .069 .180 -.050 .364 .220 -.750 -.181 -.212 .355
11 .104 -.186 -.112 -.313 .150 -.003 -.101 .060 -.278 .563 .643

Page | 21
 EXPLORATORY FACTOR ANALYSIS

REPRODUCED CORELATION:

OUTPUT 8

the residuals provided by the reproduced correlation matrix – That is consider the residuals
or difference between the actual correlations and reproduced correlations that stem from the
factor analysis model based on the data analyzed. When considering the “Reproduced
Correlation Matrix” retain the components generated by the model if only a few residuals (the
difference between the empirical and reproduced correlations represented in the lower portion
of the “Reproduced Correlation Matrix” output) exceed a difference of .05 between actual
correlations and reproduce correlations. If several reproduced correlations differ, you may want
to include more components.

A condition of the data of concern is when more that 50% of reproduced and actual
correlations differ by more than .05

Page | 22
EXPLORATORY FACTOR ANALYSIS

COMPONENT PLOT IN ROTATED SPACE

CONCLUSION:
The purpose of factor analysis is to reduce a large set of data into a smaller subset of
measurement variables. The factor scores tell us an individual’s score on this subset of measures.
Therefore, any further analysis can be done using factor scores rather than the original data.
Secondly, factor scores may be appropriate to use for Multiple Regression analysis because they
are produced from uncorrelated factors. Thus these scores reduce or eliminate multicollinearity
that can cause problems with multiple regression analysis.

Page | 23