Reidar Conradi and Finn Olav Bjørnson

Possible template for a PhD article thesis

IDI, NTNU29 October 2007

www.idi.ntnu.no/grupper/su/publ/ese/phd-thesis-template.docComments appreciated; send to Reidar.Conradi@idi.ntnu.no!

Background: This is a possible but not mandatory template for how to structure and format a PhD thesis, being a “light-weight” article thesis, not a full monograph. That is, a thesis containing an introduction and summary of some 50 pages, followed by a large Appendix with a verbatim copy of the selected and published papers. For the secondary papers, just list their authors, title, publication channel etc. and put their abstracts in another and much slimmer Appendix.

Then some admissions: there is no fixed and agreed-upon rule for how many and what kind of papers (journal, conference, …) that will suffice for an article thesis. That will depend on the traditions of the actual discipline, such as medicine, mathematics, sociology, or informatics. For the latter, the total number of selected papers will typically be 5-7. Most papers should have yourself as the prime (first) author and with at least 2 papers in journals or top-level conferences (say with acceptance rates less than 20%). 1-2 of the papers may be submitted, but still awaiting decision wrt. publication, although they are “ready for publication”. For all co-authored papers, you must briefly describe your own contribution by its share of substance and effort. Hint: Pure internal reports are discouraged. Medicine at NTNU also has a rule that co-authored papers cannot be included in more than two theses. Try to sort all this out in close dialogue with your advisor.

The rest of this page contains information regarding formatting of text and can be deleted once you’ve finished reading it. For more recent information on formatting, check the website of NTNU-trykk: http://www.ntnu.no/ntnu-trykk/ This template is formatted according to the specifications given on this site as of 25 October 2007: Top and side margins must be 30 mm, bottom margin is set to 35 mm. Normal text is formatted as Times New Roman 12pt. The reason for using 12pt is that the publisher will down-scale the A4-format of your present manuscript to a B5-format (17% linear scale-down or roughly 10pt) in the printed thesis.

Red text is explanatory “meta-“information and should be deleted from the final thesis.

Blue text is filler text, usually containing later thesis contents, and needs to be replaced.

Green text is example text and can be deleted or changed.

Title

Author

(this page will be removed by NTNU-trykk and replaced with a uniform template, the page is known as the title page)

Doctoral Thesis

Submitted for the Partial Fulfilment of the Requirements for the Degree of

philosophiae doctor

Department of Computer and Information ScienceFaculty of Information Technology, Mathematics and Electrical EngineeringNorwegian University of Science and Technology

Oct. 29, 2007

(This page will also be removed by NTNU-trykk and replaced with a template. This page is known as the colophon page, or kolofon side)

You will need the two ISBN numbers and the internal NTNU “thesis number” that year. All this can be obtained from http://ojapp01.itea.ntnu.no:7780/isbnprovider/start.do

The ISSN serial number is the same for all doctoral theses at NTNU, and is 1503-8181.

Copyright © 2007 Author

ISBN 978-82-471-48xx-x (printed version)

ISBN 978-82-471-48xx-x (electronic version)

ISSN 1503-8181

Thesis at NTNU:xxx

Printed in Norway by NTNU-trykk, Trondheim

Dedication

(optional)

Abstract10-15 lines on motivation, 10-15 lines on approach, 3-5 research questions (named RQ1, . ..), 3-5 main contributions (named C1, ...) being briefly coupled to the RQi. Start with the abstract ASAP! Keep the abstract to 1 page. ¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨¨Example from Bjørnson’s thesis stands below (in 11pt).

Reports of software a development projects that miss schedule, exceeds budget and deliver products with poor quality are abundant in the literature. Both researchers and the industry are seeking methods to counter these trends and improve software quality.

Software Process Improvement is a systematic approach to improve the capabilities and performance of software organizations. One basic idea is to assess the organizations’ current practice and improve their software process on the basis of the competencies and experiences of the practitioners working in the organization. A major challenge is to create strategies and mechanisms for managing relevant and updated knowledge about software development and maintenance. Insights from the field of knowledge management are therefore potentially useful in software process improvement efforts to facilitate the creation, modification, and sharing of software processes in any organization.

In the work presented in this thesis, we have made an overview of empirical studies on the effect of knowledge management in software engineering. We have categorized these studies according to a framework and we report findings on the major concepts that have been investigated empirically, as well as the research methods applied within the field. We have also followed software process improvement initiatives in three companies through action research studies. We examined socialization through a mentor program, and codification of software process through two approaches, one based on the Rational Unified process and one using Process Workshops. Finally we have suggested a revised method for project reviews, which we have shown empirically in a controlled experiment to be more effective than previously suggested methods for our chosen context.

We have classified the work in this thesis within three main themes: RT1: Overview of previous research on knowledge management in software

engineering.RT2: Application of knowledge management to improve the software process through

codification of knowledge.RT3: Application of knowledge management to improve the software process through

sharing of knowledge from person to person

The main contributions are:C1: An overview of the research literature on empirical studies of knowledge

management in software engineering.C2: A method for tailoring the Rational Unified Process to the development process of

a software consulting company.C3: Improvements of the Process Workshops method by contextualization.C4: Improvement of the root-cause analysis phase of the lightweight Post Mortem

Analysis for more effective project reviews.C5: Proposed methods to increase the learning effect of mentor programs in software

engineering.

i

ii

Preface

This thesis is submitted to the Norwegian University of Science and Technology (NTNU) for partial fulfilment of the requirements for the degree of philosophiae doctor.

This doctoral work has been performed at the Department of Computer and Information Science, NTNU, Trondheim, with NN as main supervisor and with co-supervisors NN1 and NN2.

A section on how the thesis was financed. If you are financed by an external project, make sure to mention it. Also give a possible project number from the Norwegian Research Council or from similar, external funding agencies.

iii

iv

Acknowledgements

Thank everyone you want to thank: advisors, colleagues, companies, friends, family etc. Try to keep it short – less than a page.

v

ContentsAbstract........................................................................................................................i

Preface........................................................................................................................iii

Acknowledgements.......................................................................................................v

Contents......................................................................................................................vi

List of Figures..........................................................................................................viii

List of Tables............................................................................................................viii

Abbreviations.............................................................................................................ix

Introduction.................................................................................................................1

Problem Outline 1Research Context 1Research Questions 1Research Design 1Papers 2Contributions 3Thesis Structure 3

State of the Art............................................................................................................4

SoA Area 1 4SoA Area n 4Research Methods in “Software Engineering” (or likewise) 4

Context and Research Design....................................................................................5

Research Goal 5Research Process 5

Study1: (etc. with StudyN)...................................................................................5Research questions and methods..........................................................................5

Results..........................................................................................................................6

Summary of the studies 6Study 1: (etc. for StudyN).....................................................................................6

Overview of the contributions 6Contribution C1: …..............................................................................................6

Research questions answered 6RQ1: ….................................................................................................................6

Paper Abstracts 6P1: …....................................................................................................................6

vi

Evaluation and discussion of results.........................................................................7

Evaluation of research questions 7Evaluation of Contributions 7Evaluation of Validity Threats 7Reflections on the research context 7

Conclusion...................................................................................................................8

Contributions 8Future Work 8Concluding Remarks 8

References....................................................................................................................9

Glossary.....................................................................................................................10

Appendix A: Selected papers...................................................................................11

P1: ….. 11

Appendix B: Secondary papers...............................................................................12

SP1: ..... 12

Appendix N: …...........................................................................................................13

vii

List of FiguresFigure 1: Example studies vs. contribution vs. papers......................................................2

List of TablesTable 1: Example of relations............................................................................................3

viii

Abbreviations

NTNU Norwegian University of Science and Technology++++

ix

Introduction

(6-10 pages) This chapter should contain: Motivation. Context e.g. a R&D project. Problem definition. Research questions. Summary of main results/contributions. Clarification of possible contributions from co-authors. Outline of rest of thesis. We suggest the following subchapters:

Problem Outline

Motivate the thesis, why is your topic an interesting one?

Research Context

In what context have you been working? Was it an externally financed project, did that put restrictions or directions on your research directions?

Research Questions

Name your research questions, outline briefly why these are interesting. Do not go too in depth here, remember it is only an introductory chapter they will be covered more thorough in later chapters.

RQ1: …RQ1.1: …+++

Research Design

Outline how your research has been happening, figures, timelines and tables that connect papers, research questions and studies are a nice help here to create an overview of the work – see figure on the next page.

1

Figure 1: Example studies vs. contributions vs. papers [Bjørnson’s PhD thesis]

Papers

If your thesis is an article thesis, provide a list of your included papers here with full bibliography. If you want to go more in depth here you could include abstracts, identify their relevance to the thesis and your contributions towards them. Keep in mind you are still in the introduction chapter and you should keep it as short as possible.

P1 Finn Olav Bjørnson and Tor Stålhane: "Harvesting Knowledge through a Method Framework in an Electronic Process Guide", Proc. 7th International Workshop on Learning Software Organizations (LSO), Kaiserslautern, Germany, 2005, 107-111 (Post conference proceedings printed in Springer LNAI 3782, 2005, 86-90)Relevance to this thesis: This paper presents our initial findings in study 3, and details how they envisioned their knowledge sharing project. It describes a tool based on the preferences of the developers and input from the research literature. The paper answers research question RQ2.1 and contributes towards

2

contribution C3 and to some degree C2. The study contributes to a small degree towards research theme RT2.My contribution: This paper is the result of a cooperation in SPIKE. I performed half of the interviews during the data gathering and was responsible for performing the analysis of the qualitative data. I was the leading author of this paper.

Contributions

Identify and list your contributions in the thesis, provide a short description of each contribution.

C1: …C2: …+++

Table 1: Example of relations

Research Question Contribution Papers FocusRQ1 C1, C2 P4, P7 COTS

Thesis Structure

Briefly outline the rest of your thesis:

Chapter 2: State of the Art

Chapter 3: Context and Research Design

Chapter 4: Results

Chapter 5: Evaluation and Discussion of Results

Chapter 6: Conclusion

Appendix A: (enclosed, selected papers)

Appendix B: (basic info incl. abstracts of secondary papers)

…

Appendix N: lab results and other data, empirical tools like questionnaires, enclosures.

3

State of the Art

State of the Art

(10-20 pages) This chapter should provide an overview of the State-of-the-art/practice: Literature study. Conclude by identifying challenges.

SoA Area 1

…

SoA Area n

Describe state of the art for various fields related to your thesis. Remember to describe enough so that your readers have the knowledge to understand your work in the context of previous research, but don’t go overboard. Identify challenges in the field that your work addresses.

Research Methods in “Software Engineering” (or likewise)

It is optional if you want this chapter in the State of the Art chapter or in the next chapter. It should outline the research methods you have used and highlight strengths and weaknesses of each one. You could also briefly describe other research methods in your field and relate your research methods to these. This is only a description of the research method. Your application of these for your thesis will be described in the next chapter.

Research Plan

Context and Research Design

(5-8 pages) This chapter first summarizes the (given) context of the research, e.g. R&D project, external partners etc. It then outlines your actual research approach for all your contributions in the thesis. It should contain the challenges from ch. 2 coupled to thesis context/project/focus. Research questions/possible hypotheses vs. empirical studies, and chosen research method(s). Overall work plan with coupling to papers. We have provided some possible chapters to include.

Research Goal

This chapter should list your research questions with a complete rationale for why they were chosen and why they were considered interesting.

RQ1: repeat the question, and why it is an interesting one.RQ2: …+++

Research Process

This chapter outlines which research methods were chosen, why they were chosen, and how they were applied in practice.

Study1: (etc. with StudyN)If your paper collection consists of several studies arranging the chapter by each study is a possibility

Research questions and methodsFor some, it might be more logical to arrange this chapter by research questions than by studies. Find what suits your thesis.

Results

Results

(8-10 pages) In this chapter you present your results from your work, coming from testing/validating/exploring the theory/research-questions by empirical studies. It can be structured by contributions, research questions, or studies done. Find what suits your thesis and results. Some also like to include the bibliography of the included papers with abstract and identified contributions towards the thesis. Do not use all of the headlines below, if it leads to the same point being said over and over. Find the approach that best makes your point.

Summary of the studies

Study 1: (etc. for StudyN)….

Overview of the contributions

Contribution C1: ……

Research questions answered

RQ1: ……

Paper Abstracts

P1: ……

Results

Evaluation and discussion of results

(8-10 pages) In this chapter you assess your results. Identify your contributions. Possible theory building (establish cause-effect). Compare to other work described in chapter 2. Suggestions for improvements. Discuss construct-, internal-, external- and conclusion-validity. The major challenge in this chapter is usually which axis you want to structure your discussion around: research questions, contributions or studies. Find what works best for you and your studies.

Evaluation of research questions

If you didn’t answer these questions in the results chapter, now is the time to revisit.

Evaluation of Contributions

How does our contributions fit with the state of the art we described in chapter 2? Do they extend the field? In what way? How do your contributions compare to your research questions? Do you have your own reflections on the contributions.

Evaluation of Validity Threats

What are the major threats to our research? Mention the major threats like: Internal Validity External Validity Construct Validity Conclusion Validity

Note that you might have to discuss these separately for each study, and every validity might not be applicable depending on what research method you have used.

Reflections on the research context

Optional. But it’s often good to reflect on the (project) context of your research and how it has affected you and your research.

Conclusion

(2-3 pages) Time for the conclusion, be short and try to nail down the essence. This should usually list the major conclusions from your previous discussion. There should also be a section on possible future directions for your work in this chapter.

Contributions

Future Work

Concluding Remarks

ReferencesIt is recommended to use automated software to insert your references here and keep track of them throughout the thesis. EndNote is a good program for ms-word being licensed by NTNU. Example of formatting for: books [1], articles in journals [2], paper in a contributed volume [3], conference [4]:

[1] J.P. Bigus, Data Mining with Neural Networks, McGraw-Hill, New York, 1996.

[2] D.E. Knuth, A combinatorial optimization approach, Theoretical Computer Science 90 (1991) 1-15.

[3] A.K. Lenstra, H.W. Lenstra, Algorithms in number theory, in: J. Leeuwen (Ed.), Handbook of Computer Science, A, Elsevier, Amsterdam, 1990, pp. 673-715.

[4] A.W. Roscoe, The Intentional specifications of security protocols, Proceedings of the Ninth IEEE Computer Security Foundations Workshop, IEEE Computer Society Press, Silver Spring, MD, 1996, pp. 28-38.

Glossary

GlossaryNot just acronyms, but real definitions of important terms.

A

B

…

S

Software Engineering – An engineering discipline which is concerned with all aspects of software production from early stages of system specification through to maintaining the system after it has gone into use [cite].

T

…

Appendix A

Appendix A: Selected papers

P1: ….. Normally include the full reference list of each paper. You should also check with your previous publisher if they have any special requests for allowing a reprint of your paper in the thesis, like including original copyright notice or a thanks to the publisher. Your papers may be uniformly formatted for the thesis. However, this may take a lot of effort to go e.g. from two columns to one, so set aside some days for this.

Appendix N

Appendix B: Secondary papers

If you’re not including all your papers from the PhD study in the thesis, list the ones you excluded - each with general info and abstract only.

SP1: .....

SP2: …..

Appendix N

Appendix N: …..

Other stuff you want to include for example: collected and processed data, software, empirical tools like questionnaires and interview guides etc. – but only if needed.