Universiti Teknologi MARA. The Perception of IT Organizations Towards Software Development Methodology Adoption

Transcription

1 Universiti Teknologi MARA The Perception of IT Organizations Towards Software Development Methodology Adoption Fazilahsul ParidalHaisah Binti Mohd Ali Thesis submitted in fulfillment of the requirements for Bachelor of Science (Hons) Information System Engineering Faculty of Information Technology And Quantitative Science May 2007

2 DECLARATION I certify that this thesis and the research to which it refers are the product of my own work and that any ideas or quotation from the work of other people, published or otherwise are fully acknowledged in accordance with the standard referring practices of the discipline MAY 30, 2007 FAZILAHSUL PARIDALHAISAH BINTIMOHD ALI

3 APPROVAL THE PERCEPTION OF IT ORGANIZATIONS TOWARDS SOFTWARE DEVELOPMENT METHODOLOGY ADOPTION SHAHALAM BY FAZILAHSUL PARIDALHAISAH BINTIMOHD ALI This thesis project was prepared under the direction of supervision of the thesis coordinator, Puan Wan Nor Amalina binti Wan Hariri, Department of System Science and it has been approved by the supervisor, Puan Juliana Hamka binti Kamaroddin. It was submitted to the Faculty of Information Technology and Quantitative Science and was accepted as partially fulfillment of the requirement for degree on Bachelor of Sc (Hons) Information System Engineering. Approved by: Puan Wan Nor Amalina binti Wan Hariri Thesis Coordinator %t^ Puan Juliana Hamka binti Kamaroddin Thesis Supervisor Date: May 30,2007 Date: May 30,2007

4 ACKNOWLEDGEMENT First and foremost, praise to the Almighty "Allah S.W.T" for giving me the mental and physical strength, and confidence to complete this research. I would like to express my sincere gratitude and appreciation to my supervisor, Puan Juliana Hamka binti Kamaroddin, for her constructive ideas, criticisms, guidance and patience throughout the duration of preparing this research. She has successfully guided me to sharpen my understanding of this research. Also a special thanks to Puan Ariza binti Nordin and Puan Wan Nor Amalina binti Wan Hariri, as the coordinators for their lectures, guidance and advices towards better understanding in producing this report. I also would like to convey my special thanks to the entire respondents for their cooperation during the survey session. Special thanks to my friends at the Faculty of Information Technology and Quantitative Science, University Technology MARA, Shah Alam for inspiring and helping me in several ways. I'm really thankful to my mother, Rafeah binti Mohd Sarip and my siblings for giving the moral support and encouragement to complete this research. Always showing concern and motivate me to overcome my frustrations in the process of conducting this study. Last but not least I owe my thanks to everyone that involved directly or indirectly in the completion of this research paper. Their support is priceless and memorable for me ever. This study is possible because of all of you. Thank You IV

5 ABSTRACT This study was conducted to investigate the factors that impact the adoption of Software Development Methodology (SDM). This study uses interview and survey instrument to identify the SDM that mostly adopted by IT organizations. The survey focuses on the organizational perspective which is significant to obtain more information in relation to the SDM selection and adoption. Five factors of adoption by Roger's were identified and used to achieve findings in this research. The factors include relative advantage, compatibility, complexity, trialability, and observability. All the relevant data that is valuable to this study will be collected and analyze to reach the possible results. Then, the results will address the methodology used in software development industry, and also the factors that has greater impact to the adoption of that method. The result found that Rapid Application Development (RAD) is the method that mostly adopted by IT organizations. While relative advantage can be the dominant factor that influences the adoption of that method.

6 TABLE OF CONTENTS TITLE PAGE DECLARATION APPROVAL ACKNOWLEDGEMENT ABSTRACT TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES LIST OF ABBREVIATIONS ii iii iv v vi x xi xiii CHAPTER 1: INTRODUCTION 1.1 Introduction Research Background Research Problem Objective of the Research Significance of the Research Scope of the Research Summary 8 CHAPTER 2: LITERATURE REVIEW 2.1 Introduction Related Studies on Software Development Methodology (SDM) Adoption Software Development Methodology (SDM) Build and Fix Model Waterfall Model Rapid Prototyping Model Extreme Programming Object-Oriented Programming (OOP) 17 VI

7 2.3.6 Rational Unified Process Rapid Application Development (RAD) Incremental Model Spiral Model Iterative Development ADR Method Robust Software Development Process Agile Manifesto Cleanroom Software Engineering ICONIX MBASE Lightweight Methodology Feature Driven Development (FDD) Jackson System Development (JSD) RITE Method TROPOS Capability Maturity Model (CMM) ISO ISO ISO (SPICE) Diffusion and Adoption of Innovation Perceived Characteristics of Innovations Adoption Stages Adoption Factors by Rogers Relative Advantage Complexity Compatibility Trialability Observability Summary 36 vu

8 CHAPTER 3: RESEARCH APPROACH AND METHODOLOGY 3.1 Introduction Research Approach and Methodology Problem Formation Research Design Population Sampling Technique Data Collection Primary Data Secondary Data Data Analysis Finding Summary 44 CHAPTER 4: RESULT AND FINDINGS 4.1 Introduction Interview Results Analysis for Online Survey Analysis for Questionnaire Analysis of Organization Description Analysis for Organizations that Adopt FSDM Roger's Factors of Adoption Relative Advantage Complexity Compatibility Trialability Observability Analysis for Organizations that Adopt FSDM and Innovation The Identification of SDM Adoption The Investigation of Factors Affect SDM Adoption by Category 69 VUl

9 4.6.1 FSDM FSDM with Innovation Own Method Summary 74 CHAPTER 5: CONCLUSION AND RECOMMENDATION 5.1 Introduction Conclusions Recommendations Summary 79 REFERENCES 80 APPENDICES IX

10 LIST OF TABLES TITLE PAGE Table 2.1: The Activities Livolve in Feature Driven Development 24 Table 2.2: The Tools Used for Feature Driven Development 25 Table 4.1: Interview Result 46 Table 4.2: Number of Method Used by IT Organizations 51 Table 4.3: SDM Adoption by IT Organizations 68 Table 4.4: Factors Affect SDM Adoption by FSDM Category 69 Table 4.5: Result for Factors Affect SDM Adoption by FSDM Category 70 Table 4.6: Factors Affect SDM Adoption by FSDM+Innovation Category 71 Table 4.7: Result for Factors Affect SDM Adoption by FSDM+Innovation Category 72 Table 4.8: Factors Affect SDM Adoption by Own Method Category 73 Table 4.9: Result for Factors Affect SDM Adoption by Own Method Category 74 Table 5.1: SDM Adoption by IT Organizations 76 Table 5.2: Summary of SDM Adoption by FSDM Category 76 Table 5.3: Summary of SDM Adoption by FSDM+Innovation Category 76 Table 5.4: Summary of Factors Affecting the SDM Adoption by Category (FSDM, FSDM+Innovation, and Own Method) 77

11 LIST OF FIGURES TITLE PAGE Figure 2.1: Waterfall Model for Software Development 16 Figure 2.2: The Innovation Diffusion Process Model (Adapted from Rogers, 1983) 29 Figure 2.3: Factors Affecting IT Diffusion. (Adapted from Rogers, 1983) 31 Figure 2.4: Stages of the Innovation Diffusion Process (Adapted from Rogers, 1983) 32 Figure 3.1: Research Approach and Methodology 3 8 Figure 4.1: Total number of ftilltime employees by category (FSDM, FSDM+Innovation, and Own Method). 48 Figure 4.2: Annual sales turnover rate by category (FSDM, FSDM+Innovation, and Own Method). 49 Figure 4.3: Types of project involves within organization by category (FSDM, FSDM+Innovation, and Own Method). 50 Figure 4.4: Methodology used by category (FSDM, and FSDM+Innovation). 51 Figure 4.5: Advantages of method usage by category (FSDM, FSDM+Innovation, and Own Method). 52 Figure 4.6: Comparison for complex method by category (FSDM, FSDM+Innovation, and Own Method). 53 Figure 4.7: Comparison for reducing method complexity by category (FSDM, FSDM+Innovation, and Own Method). 54 Figure 4.8: Comparison for the need of methodology adoption by category (FSDM, FSDM+Innovation, and Own Method). 5 5 XI

12 Figure 4.9: Comparison for factors influence compatibility by category (FSDM, FSDM+Innovation, and Own Method). 56 Figure 4.10: Comparison for compatibihty factor that influence project management process by category (FSDM, FSDM+Inno, and Own Method). 57 Figure 4.11: Comparison for the acceptance of method adoption by category (FSDM FSDM+Innovation, and Own Method). 58 Figure 4.12: Comparison for level of satisfaction by category (FSDM FSDM+Innovation, and Own Method). 59 Figure 4.13: Comparison for observability aspect by category (FSDM, FSDM+Innovation, and Own Method). 60 Figure 4.14: Comparison for reasons for methodology adoption by category (FSDM FSDM+Inno, and Own Method). 61 Figure 4.15: Years taken for Methodology Innovation by FSDM+Innovation Category. 62 Figure 4.16: Sources for idea of an innovation by FSDM+Innovation Category 63 Figure 4.17: People responsible for methodology innovation by FSDM+Innovation Category. 64 Figure 4.18: Problem with FSDM by FSDM+Innovation Category. 65 Figure 4.19: Benefit of innovated SDM by FSDM+Innovation Category. 66 Figure 4.20: Objectives of SDM innovation by FSDM+Innovation Category. 67 xu

13 LIST OF ABBREVIATIONS FSDM SDM OOP RAD RUP FDD JSD CMM Formal Software Development Methodology Software Development Methodology Object-Oriented Programming Rapid Application Development Rational Unified Process Feature Driven Development Jackson System Development Capability Maturity Model XUl

14 CHAPTER 1 INTRODUCTION 1.1 Introduction The most organizational aim is the goal to perceive new work methods that will provide organizations with relative advantages over some previous method(s). Indeed, the method which is used in software development is also known as Software Development Methodology (SDM). It is able in helping people to plan, organize, and manage the software development processes. A common software development organizational problem involves unsuccessful implementation of newly acquired methodology innovation. Although the problem becomes clearer and takes time to solve, but the acceptance of this innovated technology supposedly provides significant affects that enable other innovation being made to improve it uses. The innovation processes can be viewed over various perspectives, but this study is focused on organizational view of SDM innovation and adoption. This study describes the adoption of given innovation targeted to population of user, is called innovation diffusion. It means that the communities within the population are able to do innovation to improve performance and productivity. Loch's and Huberman's (1999) defined the diffusion as the acceptance and spread of new technology in a market or user community. This research is focused on the factors that may contribute to the SDM adoption in the software development organization. The software development will perform better since the adoption of its methodology organized and implement properly. This is supported by Rogers (1995) where he stated that the stages of adoption process concerns with the knowledge, persuasion, decision, implementation and confirmation. These stages will be used in adopting those methodologies to figure out the benefits of implementing it.

15 The study assumes that the organizational purpose of methodology adoption is to improve software performance as long as organization benefits from it. This research is attempted to find the relevant user perception of the methodology adoption. The user in this context involves programmer, developer, project manager, and organization manager which are likely to lead to user perceptions of methodology adoption performance. The elaborative impact of user perceptions includes relative advantage, compatibility, complexity, trialability and observability which are as assigned by (Rogers, 1995). The application of the innovation to SDM also creates implications for helping organizations to better manage their business reengineering. With the innovated SDM, organizations are able to reorganize the development of software process to ensure the reliability and effectiveness of the software delivery itself. Thus, organizations that provide the right methodology will give them the right software products. This impact brings the organization to properly manage SDM on their way. The perception of users of a newly implemented SDM in countries such as United States, Australia, South Africa, and Finland were assessed. These countries implemented different characteristics of innovation. Templeton and Byrd (2003) used relative advantage, voluntariness, compatibility, image, ease of use, resuh demonstrability, visibility, and trialability as the innovation variables. Huisman and livari (2002) used relative advantage, complexity, compatibility, demonstrability and trialability as their characteristics. However, Mark et al., (2002) used characteristics which is similar to the one used by Rogers (1995) and did not extend to other characteristics which are relative advantage, compatibility, complexity, trialability and observability as an approach. However, in Malaysia there is no other research on this particular topic. This paper will first describe the theoretical background that brings a need for this research. A research and its associated hypotheses are expressed and viewedfi-ompast

16 researches. The methods and results of the study are then analyzed, followed by a discussion offindings,conclusions and recommendations for future research. 1.2 Research Background Software Development Methodology (SDM) involve wide varieties of terms, issues, components, factors, models and many more. As an introduction, the definition of Software Engineering (SE) is the practice of using selected process techniques to improve the quality of software development effort. This means that a methodical approach to software development resulted to fewer defects, shorter delivery time as well as better value. It is also noted that, SDM as a significant aspect in producing software can be defined as a documented collection of policies, procedures, and processes used by a development team or organization to practice software engineering. However, methodology in the context of software engineering and project management is a codified set of practices (sometimes accompanied by training materials, formal educational programs, worksheets, and diagramming tools) that may be repeatedly carried out to produce software. By defining the right methodology will help to perform correctly and identify quality software products. However, there are many aspects that an organization needs to know in terms of how to identify the right methodology. Methodology is always related to risk management. The only challenge is to select and track the methodology in an organization is just to do it wisely. By following a methodology, it will provide sufficient process disciplines to deliver the required quality for a business succeed, as long as they avoid steps that will waste time, reduce productivity, or demoralize developers. The best approach for applying a methodology is to consider that it will be able to manage risk which is obtained from the previous projects.

17 Other significant terms that should be defined is an adoption. Adoption means to take or use something, approval or acceptance, usually applied to amendments or resolutions. This is a major contribution to this topic which concerns with the adoption of SDM. Although, the term Software Development can be described as a set of activities performed to produce software products. Software development may include new development, innovation, modification, reuse, re-engineering, maintenance, or any other activities that result in software products. The suitability or otherwise of these models, or indeed the reporting of new models, for examining the intention to adopt system or software development methodologies has received relatively little research (Roberts, 1998). Technology or methodology innovations adapted to the IT industry that rarely grow for long at one institution unless they also spread among many institutions. If it is widely used and the achievement clearly identified, the various user can support each other in the long term survival SDM innovation. This is so that the widespread adoption of innovated SDM will acquire high values to the organizational and industry investment. According to Fitzgerald (1998), the arguments for using a methodology are persuasive but many organizations are resistant to change, either to using any methodology or to changing it. This means, since traditional methodologies are in placed, there still have organizations adopted the SDM. The use of that methodology is as a guideline to ensure its efficiency and effectiveness to the organization management. Even though the adopted SDM may enhance in getting better performance, it is also requires another existence of SDM to support the use of this innovated SDM. The changes of technology and the potential projects that fit to the appropriate methodology are usually the cause of the adoption of the irmovated SDM. All the terms that had been defined earlier has significant relationship with this research. The combinations of those terms provide the valuable information to this research. Focusing on the factors that are affecting the rates of Software Development

18 Methodology (SDM) adoption on IT organizations in Malaysia, the use of Rogers' factors will evidently help researchers to measure it. Additionally, factors identified by Rogers include relative advantage, compatibility, complexity, trialability and observability. All this factors will be significant in this research to investigating how far software development companies applied Rogers' factors in adopting their SDM. 1.3 Research Problem Most of software development involves disorganized activity. The software is built without much planning, and designing of the system is a combination from the short term decisions. As long as the system is running, the user's requirements are met, it is then considered as complete and can be used in appropriate circumstances. However, the system developer should work to gain trust from users to propose a system that can meet the requirements and able to achieve high expectations from the users besides, only to be used in basic conditions. Most of IT organizations have bad practice to create this high expectation. The ways they develop a system create problems in choosing the correct SDM. Even if there are best practices available, some IT companies' still face this problem. The role of SDM as a guidance that are able to bring large impacts to the software development process gives the organization to develop the system based on the existing procedures and policies in the industry. From developer perspectives, they have consideration to choose or customize the SDM because they know the overall processes that must be included in a software development. One of the SDM problems is that the development team treats the process of software development as an orderly, rational process. Fowler (2003) challenges the notion of the engineering metaphor as a basis for 'predictive' software development and proffers an alternate view of it as an 'adaptive' process that recognizes the reality of uncertainty as part of the requirements and design processes. A fiirther problematic aspect identified was that SDM fraditionally focused on technical rather than social

19 issues. From the OASIG study, it reported that the methods appeared not to work, as system developers are too technically oriented. According to the study, it mentioned that from previous few years, the types of software that are developed are changed from a more technical systems to a more user focused system (developed for users who have limited computer skills). However, the ability of traditional methods to deliver production software in a timely manner is another serious concern raised by (Fitzgerald, 1998). Another problem raised from software development methodologies is by considering that so many systems failed. According to Russo and Stolterman (1998), they have been suggested that "currently, many software developers do not follow software development methodologies when developing software system". This explained how the real software development environment occurs today. The methodology that is being adopted only as a name, but the use of that methodology is out of control. To support the above statement, Fitzgerald (1998) in his research found that there are 60 percent of respondents did not applied the development methodology, but only six percent of them have followed the methodology rigorously. The complexities of the methodology or the system itself may rise and becomes a problem to them. So, the important thing is that the methodologies adopted must fit with the needs of developers to ensure the software development processes run smoothly. The other problems that always occur in developing software also include the application of incorrect requirements. Requirements are important since the initial development process. The lack of requirements may occur because development team failed to capture the usable and accurate data while conducting an interview session. The misunderstanding concepts by users are also the problems that always arise. Other than that, development team also takes a long time to deliver the output from each phase. Because of time, the entire development processes will be diflficuh to manage and the next processes will be delayed. Therefore, on the perspective of the development costs, it will exceed because the management team failed to plan the time as well as to follow development schedule taken for each steps. This is so that, the

20 product can still be completed but with various conflicts such as cost and schedule overruns and also the lack of certain features or functions. However, the problems mentioned above in regards to the problems that occurred due to the failure of selecting the right methodology. This research conducts to look in depth what are the factors influencing the adoption of a methodology by the organization. This is because in the real world of software development industry, there are no dominating factors that can be identified as the main motivators for using a software development methodology. This theory is supported by Avison and Fitzgerald (2003) by which they said that methodologies are often contingent on the type or a size of a project, technology environment and organizational context. By doing this research, measuring the SDM adoption rate by organization using Rogers' factor may come out with the appropriate effects. Thus, there not only be able to get information about methodologies that are available, but this can also provide the organization with the useful information about the SDM adoption. The measurement will be done with the use of methodology adoption within the organization. 1.4 Objectives of the Research This research has been conducted according to its objectives. The objectives of this research are as follows: i. To identify Software Development Methodology (SDM) applied by IT organizations, ii. To investigate the perception of IT organizations towards SDM adoption using Rogers' factors of adoption.

21 1.5 Significance of the Research The significance of the research is listed as below: i. Provide development team with the important factors on selecting and adopting correct SDM in developing quality software, ii. To expose the management team with the SDM chosen that will affect the entire software processes, iii. To provide developer with better understanding, knowledge as well as to improve skills and experiences in software development environment. 1.6 Scope of the Research The scope of the research will focus on Information Technology (IT) organizations in Malaysia that involves in software development environment. 1.7 Summary The technologies growth pursued the emergence of industrial changes and challenges. However, as globally accepted that it perform to various industries as new equipments to allow the organization to compete with others. Software development industry is a large community that fully utilizes the technological usage. Software Development Methodology (SDM) is considered as new technology applied in its industry. The adoption of methodology may provide the organization with the correct procedures and policies for development processes. Even though it becomes important, there are a number of problems faced during software development process. There is no dominating factors are used as main motivators for using software development methodology, incomplete requirements, and inability to apply adopted methodology as a guideline to develop software, cost and time overruns, high number of project failures, and low number of adopting SDM are the problems that occurred. This section describes each of problems that are related to this research and focuses on the factors of adopting those SDM.

22 CHAPTER 2 LITERATURE REVIEW 2.1 Introduction Information Technology (IT) projects such as developing software are organizational investment. It requires various types of components to complete tasks. When an organization focuses on IT projects, it commits considerable time, money and resources to the projects so that it will give values to the organization in thefiiture.to be a successful organization in IT projects will be difficult without the correct Software Development Methodology (SDM) chosen, expertise and knowledge among team members. Nowadays, software products are growing with unceasing endeavor. The implementation of that software helps people in surviving their working life. It also capable to give impacts in working environment that is willing to increase their performance and makes it challenging with other technological products. However, the software products increased depends on the various factors such the way to produce products, as well as the features andfiinctionalitiesthat must be included so that the expected outcomes will be in the future. Software Development Methodology (SDM) is relatively close to producing the software. It is one of the important aspects that must be included in the software processes and also provided the developer with the sequence of steps that must be followed. There are various methodologies available for software development. A methodology is comprised of tools, techniques and methods. The correct selection of methodology plays a crucial role for the delivery of reliable and correct software products. The selected methodology must match the characteristics of the problem domain.

23 However, according to Fitzgerald (1995), many companies develop their own or adapt existing methodologies or simply do not use at all. 2.2 Related Studies on Software Development Methodology (SDM) Adoption Due to the nature of this research topic, much of the literature that has been published does not fall under the specification of this research. Approximately, the number of the literature regarding this research is quite high because it requires various data from the past research. This paper also describes the rationale and relatedness of those issues that may reflects to the findings of this research. This study found that there are critical differences among methodologies under real circumstances. It is requires to understand how methodologies are selected or adapted or how should be selected. Clearly, this research conducted to identify what are the criteria which will reflects the selection and adoption of methodologies within IT organizations. However, various literatures assign several causes and factors that will end up with adoption of a methodology. Software development methodologies (SDM) have been defined in the regulatory manner with an underlying assumption of universal applicability. However, as industrial practice suggests such Fitzgerald et al., (1996), this assumption is fiindamentally flawed. Software development projects very rarely adopt a methodology in such a rigid fashion. Patel et al., was initially introduced as a means to improve the problems of the software crisis, SDM marked serious limitations due to the lack of any disposition to flexibly adapt to the needs of specific projects, domains and organizational settings. As a consequence, significant undesired effects were generated. Fitzgerald et al., (2003) provide the effects that include lack of a political and organizational dimension, goal displacement, developers resistance to change, and inhibition of creative thinking. 10

24 According to Fitzgerald (1998), practitioners have been reluctant to adopt SDMs resulting in more than one half of them abstaining. Furthermore, it was noted that 79% of the non-adopters intend to stay that way. While there is strong belief in their potential value, where SDMs have largely failed to deliver and this is reflected adoption rates. Fitzgerald (1998) also identified a number of 'arguments' from practitioners against the use of methodologies, and detailed the 'pressures' preventing SDM adoption. Fitzgerald raised the first argument for using a methodology is persuasive but many organizations are resistant to change, either to using any methodology or to changing methodology and only one half of organizations follow any methodology. The second argument is that the importance of the actual system being developed is often conquered by the processes of the methodology. SDMs have been developed to assist software developers build systems that meet their clients needs. These methodologies are generally formalized procedures or protocols to guide development. Clearly, there are problems with software development methodologies considering so many systems fail. Russo and Stolterman (1998) suggested that many developers currently do not even follow software development methodologies when developing software systems. They also suggest that there is currently a misfit between existing methodologies and the needs of developers. Oftentimes, multiple methods are adopted and used together for the same project, but at different phases of the development lifecycle said (Khalifa and Vemer, 2000). From the other point of view. Khalifa and Vemer (2000) were found that there are two main constructs that will be the drivers of software development method usage which are process quality and facilitating conditions. It used as a proven where there is no dominating factors have been identified as the main motivators for using a software development methodology. Huisman and livari (2002) were reported in their research the relationship of various factors that contribute to the deployment of SDMs. They allocate that the relative advantage, compatibility, trailability of a systems development methodology, an individual's experience in systems development and his/her experience in systems development methodologies, management support and 11