1. Factor Analysis: Investigating Important
Aspects for Agile Adoption in Malaysia
Presenter: Jeeson Daniel J
Authors: Ani Liza Asnawi, Andrew M Gravell & Garry B
Wills
University of Southampton, UK
{ala08r, amg, gbw}@ecs.soton.ac.uk
International Conference on Agile and Lean Software Development
AgileIndia2012, Bengaluru

2. OBJECTIVE
 To identify clusters of the variables (or items) and
how they are inter-related to produce factors.
 From here, dimensions and meanings of the
variables from the questionnaire can be identified.
 Factor analysis can help to provide a summary for
data inter-relationship and places those variables
into their groups accordingly.
International Conference on Agile and Lean Software Development
AgileIndia2012, Bengaluru

3. Data Collection and Analysis
 Online Questionnaire.
 Received 207 responses.
 For factor analysis: 27 questions regarding the
adoption and completed responses equal to 88.
 Each variables has a six point Likert-type scale ;
strongly disgaree (1) to strongly agree (6).
International Conference on Agile and Lean Software Development
AgileIndia2012, Bengaluru

4. Sample Size and Strength
of Relationship
 The suitability to conduct factor analysis needs to
be checked.
 Keiser-Meyer-Olkin (KMO) – a measure of
sampling adequacy ranges from 0 to 1.
 From the data, KMO of 0.755 was obtained –
(value from 0.7 to 0.8 are good) [1].
International Conference on Agile and Lean Software Development
AgileIndia2012, Bengaluru

5. Factor Extraction
 It is one of the procedures in factor analysis.
 Involves determining the smallest number of
factors (or components) that can best represent
the inter-relations among the sets of variables.
 Here, principal component analysis is use as the
extraction method.
 In order to determine how many number of
factors are extracted, eigenvalues (or Kaiser
criterion) and scree plot can be referred to [2].
International Conference on Agile and Lean Software Development
AgileIndia2012, Bengaluru

6. Factor Extraction: Determine How Many Factors
to Retain:
(1) EIGENVALUES rules
Compone Eigenvalues Eigenvalues Eigenvalues (cumulative
nt (Total) (% of Variance) %)
7.852 29.080 29.080
1
2.534 9.385 38.465
2
1.937 7.173 45.637
3
1.638 6.067 51.704
4
1.543 5.716 57.420
5
1.182 4.378 61.798
6
1.080 4.000 65.798
7
1.028 3.809 69.607
8
0.890 3.295 72.902
9
. . . .
. . . .
0.093 0.344 100.000
27
Table 1: Eigenvalues and Total Variance Explained

7. Factor Extraction: Determine How Many Factors to Retain:
(2) SCREE PLOT
9
8 1
7
6
Eigenvalue
5
4
3
2
3 5
2 4
6 8
7
1
0
0 5 10 15 20 25 30
Factor (or Component) Number
Figure 1: Scree Plot

8. Factor Rotation and
Interpretation
 After deciding the number of factors to be
retained, the next step is to interpret the variables
that are loaded in those factors or components .
 For this purpose: the factors are ‘rotated’.
 Rotation provides a method for interpretation and
from here the interpretability of the factors can be
improved. There are two ways in doing rotation :
(i) orthogonal (varimax) or (ii) oblique (oblimin)
 Oblique rotation used in this analysis.
 Oblique rotation produces 2 matrices – (i) pattern
matrix and (ii) structure matrix.
International Conference on Agile and Lean Software Development
AgileIndia2012, Bengaluru

9. The Related Variables with Their Loadings
(Taken from Pattern Matrix)
Factor 1 Factor 2
1. Software developers have responsibility related 1. We have mixed races in our organisation/team
to the organisation’s Agile activities=0.816 (Malay, Indian, Chinese and other races) = 0.845
2. Software developers are actively involve in 2. In our organisation, we communicate in the
setting goals for Agile activities=0.805 English language = 0.810
3. In our software development, identifying project 3. In my experience, a mind set change when
scope and suitability of project is important using Agile is important as Agile practice is
when using Agile methods =0.674 different from other software
4. In our organisation we are encouraged to be methodologies=0.434
open and transparent at all levels=0.497
5. Our organisational environment is a personal
place where people share a lot of
themselves=0.564
Factor 3 Factor 4
1. In our software development, customers are 1. I see customers’ satisfaction when using Agile
actively involved in setting goals for our Agile methods=0.881
activities=0.680 2. Our software development becomes easier
2. In my experience, practitioners with because both parties (customers and
professional skills are needed when practicing developers) are working together when using
Agile=0.656 Agile=0.867
3. In our software development, customers have 3. I see Agile boost developers’ morale=0.585
responsibility related to the organisation’s Agile 4. In sotware development, Agile methods provide
activities=0.615 quicker results=0.495
4. In my experience, when practicing Agile,
customers also have knowledge of the
methods=0.556
Factor 5 Factor 6
1. In my experience, training helps the correct 1. In my experience, Agile methods are suitable for
practice of Agile methods in our organisation= - certain technologies= -0.943
0.879 2. In my experience, tools are important to support
2. In my experience, continuous learning helps the usage of Agile methods= -0.507
knowledge transfer occuring when using Agile 3. Our organisation emphasises on achievement and
methods= -0.811 goal accomplishment where aggresiveness and
winning are common themes= -0.414
(removed from the factor)
Factor 7 Factor 8
1. Our organisational environment is a personal 1. In my experience, knowing roles and
place where people share a lot of themselves = responsibilities is essential when practicing
0.614 Agile methods = 0.694
2. In my experience, I think, knowledge about Agile 2. In my experience, attitude (such as team spirit
Table 2: The Related Variables and Their Loadings

10. Meanings for the 8 factors
Interpretation and Meaning for the 8 Factors
Factor 1: Factor 2:
Developers’ Involvement and Organisational Culture and People
Organisational-related Aspect Related Aspects
Factor 3: Factor 4:
Customers’ Involvement when Benefits/Impact when using Agile
Practicing Agile methods methods
Factor 5: Factor 6:
Disagreement over the Importance Lack of Importance of Technical
of Training and Learning when and Technological Aspects when
using Agile methods in Malaysia. using Agile methods in Malaysia
Factor 7: Factor 8:
Importance of Sharing, Knowledge- Team Commitment and Clarity of
related Aspects Purpose

11. Discussion and Conclusion
 Factor analysis identified that organisational and software
developers’ involvement as the top factor important in using
Agile.
 The results also shows ‘language’ as one of the important
aspects when adopting Agile methods.
 In terms of the impact and benefits Agile can deliver, high
loadings (more than 0.8) were found in customer satisfaction and
the ease of software development as a result of collaboration
between developers and customers.
 The results from factor analysis, for the factors (or dimension)
important for Agile adoption can help software practitioners to
consider these aspects when planning to use the methods –
following their organisation and people suitability.
International Conference on Agile and Lean Software Development
AgileIndia2012, Bengaluru

12. References
 [1] H. F. Kaiser, "An Index of Factorial
Simplicity," Psychometrika, vol. 39, pp. 31-
36, 1974.
 [2] A. Field, Discovering statistics using SPSS
(introducing statistical methods series):
Sage, 2009.
Thank you!
Any questions please email to
ala08r@ecs.soton.ac.uk
International Conference on Agile and Lean Software Development
AgileIndia2012, Bengaluru