Winning Standards Platform

.793
Extraction Method: Principal Component Analysis.
Rotation Method: Varimax with Kaiser Normalization.
a. Rotation converged in 3 iterations.

The principal component analysis groups the items into 2 components. Items in
technological skills and services are being accounted in 54% of the variance and
items in platform long term viability in 18% of the variance.
Based on the factor analysis done, 6 factors are being used for multiple regression
analysis. The items are being identified accordingly to measure the underlying
dimensions. The list of factors are relative advantage, perceived compatibility &
trialability, perceived complexity, management support & knowledge expertise,
technological skills and services as well as the last factor platform long term viability.

4.5.3 Multiple Regression
The standard multiple regression was applied to test the research hypotheses. This is
due to the dependent variable which is a continuous variable and because as the
scores are normally distributed (Pallant, 2011). The regression analysis is used to

Page 65
“predict an outcome variable from one predictor (simple regression) or several
predictor variables (multiple regression)” (Field, 2009).
The first analysis is done on the correlation of the new factors with the dependent
variable In model summary Table 4-22, the overall model explains 80.3% of the
variance. The R squares explained that the model explains 64.5% of the variance in
OSS adoption.
Table 4:21: Model Summary
Model Summary
b

Model R R Square
Adjusted R
Square
Std. Error of
the Estimate
1 .803
a
.645 .627 .79751
a. Predictors: (Constant), Platform Long Term Viability, Perceived Complexity, Relative
Advantage, Technology Skills & Services, Management Support, Knowledge & Expertise,
Perceived Compatibility & Trialability
b. Dependent Variable: Open Source Adoption
To look at the significance of the relationship between the factors, the ANOVA
analysis was conducted. The results were as per Table 4-23. It is reported that the
significance value is 0.00, where there is about zero chance in 1000 type 1 error. This
also shows that the data reliability with the OSS adoption decision is strongly related.
Table 4:22 : ANOVA Table
ANOVA
a

Model
Sum of
Squares df
Mean
Square F Sig.
Regression 135.131

6

22.522

35.410

.000
b

Residual 74.415

117

.636


1

Total 209.546

123


a. Dependent Variable: Open Source Adoption
b. Predictors: (Constant), Platform Long Term Viability, Perceived Complexity, Relative
Advantage, Technology Skills & Services, Management Support, Knowledge & Expertise,
Perceived Compatibility & Trialability


Page 66
From the correlation table 4-24 below, 2 factors have high correlations with the
dependent variable. The factors are perceived compatibility & trialability, The
Pearson Correlation r value for perceived compatibility and trialability is 0.720 with
and significance at p=0.00. The other factor is management support, knowledge &
expertise resulted r=0.685 and its significance is at p=0.00.

Page 67
Table 4:23: Correlations Table

Correlations

Open
Source
Adoption
Relative
Advantage
Perceived
Compatibility & Trialability
Perceived
Complexity
Management
Support, Knowledge
& Expertise
Technology
Skills & Services
Platform
Long Term
Viability
Pearson
Correlation
Relative Advantage .387 1.000 .268 .081 .267 .256 .255
Perceived Compatibility
& Trialability
.720 .268 1.000 -.276 .711 .533 .703
Perceived Complexity -.383 .081 -.276 1.000 -.354 -.145 -.198
Management Support,
Knowledge & Expertise
.685 .267 .711 -.354 1.000 .612 .544
Technology Skills &
Services
.459 .256 .533 -.145 .612 1.000 .489
Platform Long Term
Viability
.576 .255 .703 -.198 .544 .489 1.000
Open Source Adoption .000 .000 .000 .000 .000 .000
Relative Advantage .000 .001 .185 .001 .002 .002
Perceived Compatibility
& Trialability
.000 .001 .001 .000 .000 .000
Perceived Complexity .000 .185 .001 .000 .054 .014
Management Support,
Knowledge & Expertise
.000 .001 .000 .000 .000 .000
Technology Skills &
Services
.000 .002 .000 .054 .000 .000
Sig. (1-tailed)
Platform Long Term
Viability
.000 .002 .000 .014 .000 .000





Page 68
Table 4:24 : Coefficient Table
Coefficients
a
Unstandardized
Coefficients
Standardized
Coefficients Correlations
Collinearity
Statistics
Model
B
Std.
Error Beta
t Sig.
Zero:order Partial Part Tolerance VIF
(Constant) .646

.521

1.239

.218


Relative Advantage .217

.059

.216

3.650

.000

.387

.320

.201

.868

1.152

Perceived Compatibility &
Trialability
.485

.120

.375

4.025

.000

.720

.349

.222

.349

2.862

Perceived Complexity -.208

.066

-.191

-3.168

.002

-.383

-.281

-.175

.832

1.202

Management Support,
Knowledge & Expertise
.269

.092

.259

2.929

.004

.685

.261

.161

.387

2.586

Technology Skills & Services -.029

.077

-.027

-.371

.712

.459

-.034

-.020

.580

1.725

1

Platform Long Term Viability .114

.098

.091

1.153

.251

.576

.106

.063

.484

2.065



Page 69
The coefficient analysis gives the estimates for standardizes beta. This value indicates
the relationship of dependent variable and the predictor. As per Table 4-25, relative
advantage has positive b-values (0.216) which indicate a positive relationship with
OSS adoption. This is similar to perceived compatibility & trialability, management
support, knowledge & expertise as well as platform long term viability where positive
b-values are recorded. Two items showed negative relationship; perceived complexity
and technology skills & services (b=-0.191 & b=-0.027).

Based on the significance values, it is identified that four factors are significant to
OSS adoption. They are:
1. Relative Advantage
2. Perceived Compatibility & Trialability
3. Perceived Complexity
4. Management Support, Knowledge & Expertise

4.5.4 Hypothesis Testing
The first hypothesis (H1) tested on the relationship between perceived relative
advantage and OSS adoption. As reported in the coefficient analysis above, the
standardized coefficient (β) between perceived relative advantage and OSS adoption
is 0.216 and the significance at 0.000, which is significant at p < 0.05. In other words,
there is high level perceived relative advantage of OSS adoption in the organization.
Thus, the result provides support for H1.


Page 70
The second and fourth hypothesis (H2, H4) tests the relationship of perceived
compatibility and trialability with OSS adoption. The standard coefficient beta (β) is
0.375 with significant level at p=0.000. This shows that the item is significant at p <
0.05. With this, the perceived compatibility and trialability does significantly
contribute to the OSS adoption in an organization.

The third hypothesis (H3) tests the relationship of perceived complexity with the OSS
adoption. The regression table state the coefficient beta (β) at -0.191 and it is
significantly contributed to the OSS adoption by having a significant level which is
less than 0.05 (p=0.002). Thus, it is perceived that complexity has a unique
contribution to the OSS adoption.

In the organizational context, two hypotheses were tested and during the regression
analysis, both items were grouped together. The fifth hypothesis (H5) states the
relationship between management support and the OSS adoption while (H6) relates
the knowledge and expertise with the OSS adoption. The beta (β) value is at 0.259
and the significance level is at 0.004. Hence this results shows that management
support, knowledge and expertise significantly contributed to the OSS adoption for
the significant p<0.05 and does support H5 and H6.

In the environmental context, the seventh hypothesis (H7) relates the technological
skills and services to the OSS adoption. As shown in the table above, the standardized
coefficient (β) is -0.27 and the p-value is 0.712, which is more than p at 0.05. Hence,

Page 71
the result does not support for H7 and H7 is insignificantly relates to the OSS
adoption.

The last hypothesis (H8) tested on the platform long term viability relationship with
the OSS adoption. The results in the table above, shows that the coefficient beta (β) is
0.091 and the p-value is 0.251 which is higher than p at 0.05. This means that the
platform long term viability does not make a significant unique contribution to the
OSS adoption. Hence, the result does not support H8.

Overall, two hypotheses are not supported by the results of the analysis and both of
which are under the environmental context. On the other hand, both technology and
the organizational context do have a significant contribution to the decision on the
OSS implementation in the organization.

4.6 Summary of Research Results
In total 6 hypotheses are supported and 2 hypotheses are being rejected. The summary
of the hypotheses testing results are shown in table 4-26 below.







Page 72
Table 4:25: Summary of Hypothesis Testing Results
No Hypothesis Conclusion
Technological Context
H1
Perceived higher relative advantage of OSS is positively
affecting the adoption of OSS.
Supported
H2
Perceived higher compatibility of OSS is positively
related to the adoption of OSS.
Supported
H3
Perceived complexity of OSS is negatively affecting the
adoption of OSS.
Supported
H4
Perceived triability of OSS is positively related to the
adoption of OSS.
Supported
Organizational Context
H5
Greater top management support of OSS is positively
affecting the adoption of OSS.
Supported
H6
Higher knowledge & expertise of OSS is positively
related to the adoption of OSS.
Supported
Environmental Context
H7
Higher availability of technological skills and services is
positively related to the adoption of OSS.
Rejected
H8
Greater platform long term viability is positively
affecting the adoption of OSS.
Rejected


4.7 Discussion of Research Result
As per the factor analysis that has been made, the factors have been reduced to only 6.
These are:
1. Perceived Relative Advantage
2. Perceived Compatibility & Trialability
3. Perceived Complexity

Page 73
4. Management Support, Knowledge and Expertise
5. Technology Skills and Services
6. Platform Long Term Viability

From the regression analysis done, four out of six factors are being supported and
indentified as significantly contributing to the OSS adoption in organizations. From
the four factors, the two highest significant factors for the contribution to the OSS
adoption are perceived compatibility & trialability as well as management support,
knowledge and expertise with coefficient beta (β) at 0.375 and 0.259 respectively.

Perceived compatibility and trialability of OSS adoption is one of the important
factors in deciding the OSS implementation. In this study, the respondents were asked
whether the ability to co-exist with the current infrastructure and processes as well as
the ease of use and testing significantly contributed to the adoption. Supported with
previous studies, this factor is significantly important to the organization who adopts
OSS (Dedrick & West, 2003, 2004; Glynn et al., 2005; Kshetri, 2005; Morgan &
Finnegan, 2007; Ven & Verelst, 2008; West & Dedrick, 2005).

Management support, knowledge and expertise in the organization are also important
as to support the usability of the system. This is supported by Goode (2005) study
which reads that the management would be unwilling to explore the extant of an OSS
should there be no business need, thus contributing to low support from the
management. Ven and Verelst (2008) stated in their study that the organization that

Page 74
cannot easily obtain both internal and external knowledge are less likely to adopt the
OSS. Quite a number of literature discussed the existence of boundary spanners in the
organization which also leads to the availability of OSS expertise in the organization
(Dedrick & West, 2003, 2004; Glynn et al., 2005; Gurusamy & Campbell, 2012; Ven
& Verelst, 2012).

Apart from these two factors, the other factors which also significantly contributed to
OSS adoption is under the technology context, relative advantage (β= 0.216) and
perceived complexity (β= -0.191).

Perceived complexity and perceived relative advantage are among important factor in
the adoption decision of an OSS and the organization who perceives OSS as complex
is less likely to make use an OSS (Ellis & Belle, 2009). The finding of relative
advantage in the implementation cost (both hardware and software) is consistence
with previous research (Dedrick & West, 2003, 2004; Glynn et al., 2005; Larsen,
2004; Lundell, 2006; Morgan & Finnegan, 2007) so is the incompatibilities of OSS to
the existing infrastructure causing the organization to incur switching costs (Dedrick
& West, 2003; Fitzgerald, 2003; Ghosh, 2005; Goode, 2005; Ven & Verelst, 2012) .

The findings also indicate that the OSS implementation is not complex and this is also
being supported by Chau and Tam (1997) in their study. The adoption of a complex
technology can be described as a process of accumulation and is not an ‘overnight’
event.

Page 75

However, two factors in the environment context were not supported; technological
skills and services and platform for long term viability. These two factors were
strongly supported by previous studies (Dedrick & West, 2003; Ven & Verelst, 2012)
but somehow they are not supported in this study. However, not all contexts
discussed in this study have a direct impact to the adoption, which are also identified
as depending on which the adoption takes place (i.e. different region), would have
different results (Ven & Verelst, 2012)
.


4.8 Chapter Summary
This chapter presents the research results and the analysis. The data is analyzed based
on descriptive statistics of the respondent’s profiling as well as the reliability testing
using Cronbach’s alpha analysis, test of normality and correlation analysis using
Pearson’s correlation co-efficient. The data was then further analyzed using standard
multiple regression to obtain the significance level as well as measuring the strength
of relationships between the factors and the OSS adoption.

Page 76
Chapter 5 Conclusion and Recommendation
5.1 Introduction
This chapter covers the summary of this study. It will start with a summary and
conclusion as well as discussion on the limitation of this study. There will be
suggestion for future work for other researcher and implications are also discussed.

5.2 Summary and Conclusion
This research uses the TOE theory of Tornatzky and Fleischer (1990) framework as a
foundation to study the adoption of OSS from the Malaysia perspective. Three
contexts were being investigated to understand the decision to adopt the OSS. The
three contexts are technology, organization and environment.

Factors for each of the context were identified based on the previous studies done on
OSS adoption and thus, data collection was performed based on a survey. In total,
124 organizations responded to the survey and the data was analyzed and grounded in
the technology adoption literatures. The technology and organizational contexts were
found to be significantly important in the OSS adoption in Malaysia. Four factors out
of six were identified as the most important factors to the OSS adoption namely
“perceived relative advantage”, “perceived compatibility and trialability”, “perceived
complexity” and “management support, knowledge and expertise”.


Page 77
Based on the findings above, in Malaysia, there is a high level of OSS systems
adoptions in the organization especially in the Government sector. This finding is
supported by the initiative by the Government on the implementation of OSS systems
in the public sector. Malaysian Administrative Modernization and Planning unit
(MAMPU) has the capabilities of developing open source applications that can be
used for free by all government agencies as well as various improvements have been
made to the existing systems. Thus, with such efforts made, many small organizations
and businesses have able to benefit form it.

This research also attempts to identify the variables and significant factors that relate
to the OSS adoption as well as the level of the adoption. Thus the first question
addressed in this research is “What is the level of adoption of OSS enterprise systems
in Malaysia?”. Based on the results, the level of adoption is high especially in the
government sector in Malaysia. As mentioned in chapter 4, based on the results, 73%
of the respondents were from the government sector and another 27% were from
various industries. The results also show a very high implementation in operating
systems as well as the databases.

Research was then further conducted to analyze the significant factors that influence
the manager’s decision by constructing the second question “What are the significant
factors that influence a managers’ adoption of OSS enterprise system”. This question
fulfills the first three objectives of the research by using the TOE framework as the
basis.

Page 78

The research adapts the Tornatzky and Fleischer (1990) framework and it uses the
grounded theory developed by Dedrick and West (2003) to identify the factors for
each of the contexts in the TOE framework. The first context that this research used is
technological. In technological, four factors were being used: relative advantage,
compatibility, complexity and trialability. From these four factors, this research
managed to reduce the four factors to three factors. The three factors are perceived
relative advantage, perceived compatibility and perceived triability and perceived
complexity. All three factors were found significant to the adoption of OSS.

The second context this research focused on is the organizational which fulfilled the
second objectives of this research. Two factors were being identified, management
support as well as knowledge and expertise. During the factor analysis, these two
factors were reduced to one i.e. “management support, knowledge and expertise”.
The results show that this factor is significant in relation to the OSS adoption. It is
therefore safe to assume that it is important for any OSS implementation to have
support from the top management as well as having internal knowledge and expertise.
This is to ensure the success of the implementation. It is noted that support from
management is because of the low cost that is associated with OSS implementation
(Morgan & Finnegan, 2007).

The last context discussed in this study is the environment. Two factors were being
identified; technological skills and services and platform long term viability. These

Page 79
factors were being highlighted as important factors in adopting OSS in numerous
studies but it have been proven not to be supported in this study. In Malaysia, these
two factors do not influence the adoption of OSS in the organization. It could be due
to the fact that Malaysian organizations consider other factors which are more
important such as the monetary aspect or the relevance and requirements of the OSS
to the business.

Without the technological skill and services, the organization would opt to have its
own internal staff to support the system which is also relates to the availability of
knowledge and expertise in the organization in the organizational context. This could
be one of the reasons why technological skills and services factor was rejected in
relation to the OSS adoption.

Similarly, for platform long term viability, the results showed that this factor was
rejected in relation to the OSS adoption. As mentioned earlier in this study and
defined by Coppola and Neelley (2004), the improvement of OSS are being
contributed by the users or the community who have fixed or added new features to
the software. Thus, it is not required for the OSS to be winning standards or enriched
features in order for the user to implement the OSS.



Page 80
5.3 Limitation to the Study
Based on the data, this study has several limitations that affect the generalization of
the findings.

The first limitation is due to the unfairly distribution between the public and private
sector and this is due to several of reasons:
1. Low response of the survey from the private sector as there is no specific
list of such organization which implements the OSS in Malaysia. The
distribution of the survey was by sending out email to the OSS
community forum as well as by using the social media. Using this
method, the target respondent cannot be reached.
2. During the data collection, there was a conference on Government OSS
held (MyGOSS 2012). But the researcher was unable to participate in
the conference due to the conference organizer’s disapproval of the
researcher’s participation in the conference, hence contributing to the
low data collection.

A second limitation to this research is the collection method of this survey. The main
method of distribution is via email and the distribution list is only on the public sector
based on the OSS adoption report by OSCC. Due to time constraints, the target
respondents were being contacted via email. Thus, the response will be based on the
recipients’ decision on whether or not to participate with the survey. This may limit
the researcher to generalize the findings.

Page 81

A third limitation is due to the location of the organization, which is unfairly
distributed where most of the organizations who responded were located in the Klang
Valley and Putrajaya. Only a few of the organizations are from other states in
Malaysia which do not represent the overall findings.

Lastly, this study was taken at the point of time of the OSS implementation. The
results might differ by doing surveys on continuous usage or cross sectional surveys
of the OSS implementation.

5.4 Suggestions for Future Research
It would be interesting to do an exploratory research of this study to investigate
further the actual perceptions of managers on the OSS adoption. This is to address
any other factors that were not counted such as the security features of the systems
and source code availability.

Further study can also be done to differentiate the effects in the public and private
sector separately as this study is generalized for both sectors. The significant factors
might be different between the two sectors.

This study also can be extended to different levels of respondents. Data can be
collected from the top management level in an organization to the end users. This is
to explore different views of the OSS adoption in an organization.

Page 82

Data can be collected from other states in Malaysia to generalize the findings. In the
current study, data collected was mainly from the Klang Valley, Putrajaya and a few
from Pulau Pinang, Pahang, Sabah and Sarawak.

The external support in the market could also use this research to create a better
support and services to be offered to the organization. This research can also be
duplicated by other practitioners in other regions to test out the effects especially in
developing countries.

5.5 Implications
This study provides empirical study of the technological, organizational and
environmental factors in Malaysia. Thus, giving managers some insight before
adopting the OSS enterprise system.

This study shows that from the TOE framework, only two contexts are applicable in
the Malaysia perspective. The contexts are technological and organizational. This
shows that a different finding may result when conducted in a different region than
the originated study. The findings from this study add evidence to existing studies on
the OSS adoption specifically to those using the TOE framework.

For practical use, the findings from this study assist managers to concentrate on the
significant factors for an OSS implementation in their organization.

Page 83
• Managers should consider the relative advantage when evaluating the OSS
application to the organization. In this study the respondents agree that the
cost of implementation for hardware, software and switching cost were
significant in adopting the OSS.
• Compatibility and trialability were also cited in various study of the OSS
adoption (Dedrick & West, 2004; Glynn et al., 2005; Gurusamy & Campbell,
2012). The respondents preferred more compatible OSS platform to the
existing applications to ease the transition process. Organizations would also
like to be able to test the software before implementing it. And the same
applies to the complexity factor. Similarly, this study has proven that these
factors are significant with the OSS adoption.
• Top management support has proven to be a crucial factor in implementing
the OSS. Thus, decision makers should ensure the ‘buy-in’ from the top
management for a successful OSS adoption to the organization.
• The availability of internal knowledge and expertise is also another factor to
consider when evaluating the OSS. Without internal knowledge and expertise,
it is harder to adopt the OSS implementation in the organization.