Integration of Agile Practices into Traditional SoftWare Development Life Cycle - Research Proposal Example

Integration of Agile Practices into Traditional SW Development Life Cycle Introduction i) Background of the Study Software development projects are complex thus requiring software development life cycle. The life cycle is a systematic process of building and maintaining software systems (Leau et al., 2012). This process has evolved greatly in the past few years with an experienced growth in software technology. The traditional software development life cycle involved the traditional software development methodologies, which are Waterfall methods, V-Model, and Rup (Okoli & Carillo, 2010). Such methodologies had many shortcomings when used in the development of a software system thus leading to the introduction of modern software development methodologies, popularly known as agile practices. According to Leau et al. (2012), agile practices include Crystal methodology, dynamic software development method, feature-driven development, lean software development, scrum, and extreme programming. These practices could be integrated into the traditional software development life cycle to change how project are run instead of completely changing to the new methods. This would enable software developers take advantage of both methodologies. i) Purpose of the Study The purpose of this study is to identify how the agile practice could be integrated into traditional software development cycle instead of completely changing the cycle. The integration can be done in various ways whose difference depends on the type of practice. In the study, we will identify several agile practices as well as analyze how each one of them could be integrated to the cycle to achieve the purpose of the study. ii) Delimitation and limitation The study will have some limitations and delimitations based on the limited resources available for the research and the dynamic changes in the technology of software development. We will therefore use qualitative approach due to the limitation in the resources available for the research. This approach is economical and does not require many resources. Moreover, because of the dynamic nature of the agile practices, we will not be able to study all of them. In this study, we have planned to discuss only the above-mentioned six agile practices although there are some that have lately introduced. Another limitation of this study is the many ways in which agile practices could be integrated into the traditional software development life cycle. Because of these, we will not be able to cover all of them in this study. What will be covered in this study will be the main ways in which these integration could be done. iii) Significance of the Study Traditional as well as modern software development life cycle have both strengths and weaknesses. This means that none of them is sufficient on its own for achieving effective software development. Integration of agile practices could enable software developers to take advantage of the strengths of both the traditional and modern software development life cycles. It would also enable software developers to avoid the shortcomings of each of the cycles. The software development approach that would be generated through this integration will therefore be sufficient in achieving effective software development. That is why this study is very significant and worth the time and resources it will consume. Research Methodology i) Theoretical framework For this research, we will use a case study qualitative research method. This is a type of a research method brings together aspects of a case study and aspects of a qualitative research. Case study studies the complexity and particularity of a single case (Stake, 1995). Since for this study we will need to understand the particularity of software development life cycle, then we would need to use a research method with an aspect of case studies. According to Travers (2001), this kind of approach involves data aspects like positivism, intepretivism, realism, and post-structuralism. These are what determines the design of the research and how the research will be used to answer the research question. For this research methodology, none of the forms of numerical data is used to draw conclusions. Case study qualitative research method has several advantages over other research methods. One of the key advantages of this method is they it has very few assumptions. Having few assumptions increases the degree of accuracy of the analysis process. Another advantage of this method is that it involves less procedures thus it is not time consuming. When this approach is used in research, the three key stages of research, which are data collection, analysis, and interpretation, can be taken within a short period. Since we do not want our research to take long before we can be able to draw conclusion, this is the most appropriate approach to use. However, this kind of approach to research has several disadvantages. One of the disadvantages is that in this kind of approach, we cannot undertake inferences. This means that when this methodology is used, data should be collected on the whole population. Otherwise, it would not be possible to draw conclusion about the population. Moreover, there are very few statistical software systems that have been generated to undertake analysis of research undertaken using this methodology. Furthermore, the degree of accuracy of the developed software systems could not be trusted in terms of accuracy. This means that for this kind of research methodology, we have to rely on non-computerized analysis that is tedious. i) Selection and description of participants Participants who will make the population of study for this research will be software developers. Due to limitation in the resources and time available for the research, it will not be possible to use the whole population. We will therefore, select a sample of software developers from the whole population. Since for this methodology, inference could not be done, the sample selected should be large enough and sufficient. This would enable it to provide us with enough information about the issue we would be studying. As the research team, we thought that thirty developers would be a good number for our sample as long as they will be well selected from the population. Since there will be no inference in this research method, there will be no need of using simple random sampling to select our sample. Furthermore, there is no need of ensuring that the sample we come up with is representative of the whole population. The thirty participants identified for our sample will come from the five most popular software development companies. These companies will be identified in terms of the number of software systems they have developed in the last one month. As the research team, we believe that by selecting the companies using this criterion we would come up with companies that applies both modern and traditional software development life cycles in their production of software systems. For a software developer to qualify for inclusion in our sample, they will be required to have several qualities. One of the qualities is that he should have at least a ten years’ experience in software development. This will ensure that we have people who are familiar with both software development cycles. A second quality is that he will need to be in a senior position in a software development team. A software developer in a senior position must have felt the effect of both software development cycles when used to develop a software system. This means they would have selected a person who is familiar with the main issue in our study. However, even after satisfying these two qualities, the participant must be willing to provide the required information. ii) Data Collection strategies Data collection strategies adapted in a research depends mainly on the research method used in the study. The strategies that are appropriate for a quantitative research method might be very inappropriate for a qualitative research method. According to Travers (2008), the strategies that could be used for a qualitative research are observation, interviewing, ethnographic fieldwork, discourse analysis, and contextual analysis. Each of these strategies has its own advantages and disadvantages over other strategies. Moreover, each of them has specific situations in which it could be the most appropriate. However, there are some that could be used together to supplement each other especially in a case where there is a need to collect a lot of information. For this specific situation, we will use observation and interviews as our main data collection strategies. After the participants have been identified for the research, an observation will be done on the work they have done. The other data that will be on the visible attributes that might be influencing their software development work. Majority of the data in this research will be collected through conduction an interview on the participants. The interview will contain questions related to how we could integrate agile practices on traditional software development life cycle to make it better. The questions used to collected data in the interview will not be many to avoid consuming a lot of time in administering the interview. iii) Data analysis strategies Data analysis strategies that would fit a specific research also depend on the type of a research. For a qualitative research like this one, qualitative data analysis strategies would be the most appropriate. This kind of analysis involves non-numerical observation of information collected throughout the study. Like in quantitative analysis, the data collected in this case will need to be processed before the actual analysis. The methods that could be used to process qualitative data are coding, memoing, and concept mapping (Babbie, 2010). For this research, we will mainly use concept mapping to process our data so that we can make the process of analysis and interpretation much easier. To analyze the data collected in this research, the main techniques that will be used will be examination and evaluation. The data collected will be examined to identify how much they fit the issue studied as well as the solutions they suggest. Collected data will also be evaluated to determine the impact, consequences, and applicability of what they suggest. Moreover, some data will need to analyze using categorization. Data collected through observation will be analyzed using this qualitative data analysis method. These data will not be part of the main analysis, but they will be very crucial for the interpretation process. Literature Review Several scholars have studied these two software development life cycles. All of them indicate that their procedure and sequences of activities are different. Leau et al (2012) found out that the traditional software development life cycle starts with the design phase, then the coding phase followed by the testing phase. All the three phases are interconnected and they must follow the order indicated above. Stamelos & Sfetsos (2007) found that the modern software development cycle involve small iteration cycles that are interdependent. In this kind of software development cycle, each of the steps involved depend on its preceding step. Moreover, the software system formed is provided to users and stakeholders in each of the steps for comments and suggestions. Moreover, studies have already been done on the advantages and disadvantages of these two types of software development cycles. Leau et al (2012) found that traditional software development life cycle is time consuming and does not produce software systems well suited for the customers’ needs. This is because they do not involve customers in the development stages. Moreover, in the initial phases the software are not implementable and thus consumers would not be able to make comments about them. However, Okoli & Carillo (2010) found out that this software development life cycle has an ability to manage large projects, attention to quality assurance, and long-term scalability. This means that it should not be completely discarded for the agile software development life cycle. However, scholars have found the agile software development life cycle to have several qualities that are not in the traditional software development cycle. Leau et al (2012) found out that the type of software developed using this form of a life cycle suite specific customer needs. Therefore, software systems made using this cycle do not need adjustments after they have been released to the customers. However, this software development cycle is only suitable for small projects with limited set of characteristics (Okoli & Carillo, 2010). This means that they cannot be sufficient on their own since in most cases software development involve large complex projects. Timeline Feasibility Four weeks will be needed to completely conduct this study and write a report of our finding. The first step of this study, which is the preparation stage, will only take a few days. This is because it will not involve many activities. However, the data collection stage is expected to consume most of the time allocated for this study. We expect it to take a minimum of two weeks and a maximum of three weeks. This is because it involves many activities and is thus the major activity in the study. The analysis, interpretation and reporting will consume the remaining one week and thus after a month the result of the study will be available. References Babbie, E. R. (2010). The practice of Social Research. Belmont, CA: Wadsworth Cengage Learning. Leau, Y. B. et al. (2012). Software Development Cycle Agile vs Traditional Approaches. Retrieved From: http://www.ipcsit.com/vol37/030-ICINT2012-I2069.pdf. On 13 August 2012. Okoli, C., & Carillo, K. (2010). The Best Adaptive and Predictive Methodologies: Open Source Software Development, a Balance between Agility and Discipline. Retrieved From: http://chitu.okoli.org/images/stories/bios/pro/research/open/OkoliCarillo2010IJAESD.pdf. On 13 August 2012. Stake, R. E. (1995). The Art of Case Study Research. Thousand Oaks: SAGE Publications. Stamelos, I. G., Sfetsos, P. (2007). Agile Software Development Quality Assurance. London: Information Science Publishers. Travers, M. (2008). Qualitative Research through Case Studies. Thousand Oaks: SAGE Publications. Read More