The final step in the research process is to put all the pieces together in a cogent conclusion. identifying limitations associated with those conclusions, and recommending the future research questions and studies. How will you draw conclusions?  Are there limitations that you have identified in the design thus far?  The researcher makes many decisions during the research design process that directly drive some of the limitations.  It is helpful to identify these limitations when the design decisions are being made.

Expected Contribution to Knowledge

With the conclusions drawn what do you expect will be the contribution to knowledge?  This should directly address the purpose statement.

Expected Outcome

Finally, what is the expected outcome of the study?  This should directly link to the problem statement. What will managers be able to do with this new knowledge?

Example - Chad McAllister’s [Prospectus Version]

Drawing Conclusions

Conclusions will be drawn principally from three areas:

1. The weighted factors that influence misunderstandings of requirements.

2. The differences in factors and their weightings between users and developers.

3. The similarities and differences in definitions of critical factors between users and developers.

Insights are expected to be gained about why users and developers misunderstand requirements.  Limitations of the study are largely due to its exploratory nature and include:

1. Relative small sample size that may not be representative of IS development in other companies.

2. Variations in software development processes may impact the factors users and developers would identify.

3. No correlation of factors to successful versus unsuccessful IS projects is attempted—factors may differ based on project success.

4. The study focuses on misunderstandings of requirements between users and developers.  Although these are two key groups in the development of software, influences of other stakeholders are not controlled for.

5. No attempt is made to associate misunderstanding of requirements with the quality of the software.  Although many studies correlate well understood requirements with software quality, minimizing the factors responsible for misunderstandings does not directly imply that the quality of software will improve and such a correlation would need to be tested.

6. The factors identified are from an IS domain and may or may not be the same factors that would be generated from other domains, such as commercial software development or new product development in the manufacturing domain.

Follow-on research opportunities are in part driven from the limitations and include:

1. Using the NGT [nominal group technique] approach with other companies to examine the completeness of the misunderstanding factors.

2. Conducting a survey of users and developers drawn from a random sample to weight the factors identified in this study to promote the generalization of the results.

3. Examine how the factors vary based on different software development methodologies, such as Waterfall, Iterative and Incremental Development, and Extreme Programming; or based on the success of the software project as viewed differently by developers and users. 4. Propose methods for minimizing the critical factors that influence requirement misunderstandings and assess if the quality of requirements and/or the information system improves.

5. Critically examine one or more popular methods for improving the understanding of requirements, such as VOC, in light of the factors discovered here.

Expected Contribution to Knowledge

As exploratory research, the study will lay a foundation for further work that could show a correlation with minimizing misunderstandings of requirements and quality of software.  By knowing the factors that influence misunderstandings of requirements and the different perspectives between users and developers, methods could be proposed and tested for improving the understanding of requirements.  Such improvements are expected to increase the quality of information systems.

Expected Outcome

By knowing why requirements are misunderstood we will be are better prepared to devise ways to improve users' and developers' understanding of requirements. Although many methods have been proposed for this, such as VOC, a theoretical knowledge of the factors responsible for misunderstanding is lacking.  With knowledge of the factors, enhancements to VOC and other methods, as well as the creation of new methods, can lead to more effective and efficient requirement engineering processes. An immediate outcome of the study is the ability of software project managers to minimize misunderstandings by minimizing the factors that influence misunderstandings. Project managers will be better informed to identify factors on their development projects that are jeopardizing the projects success. 

Example Stock and Flow Model - 1

Click on diagram for larger PDF view.

Example Stock and Flow Model - 2

Click on diagram for larger PDF view.

For more on stock and flow models see Jay Forrester’s work on Industrial Dynamics.

Align and Integrate

1. The approaches to drawing conclusions should provide the new knowledge and insights needed to help solve the problem.

2. The conclusions should be derived from and consistent with the data analysis methods.

3. As with all the components of the research methodology the conclusions should be appropriate for the variables, relationships, context, etc. identified in the conceptual framework.