You have a working prototype, forty pages of notes and a supervisor who returned your chapter outline with one comment: “This is not how an IS dissertation is structured.” Registration for next year falls due in a few months, and the difference between submitting in time and paying another year of fees is usually not the artefact or the survey; it is whether the chapters are in the order the department expects. Here is that order, for both traditions of South African information systems research, with what each chapter must contain and where students lose the most time.
Build your information systems chapter outline in Tesify, free
First decide which of the two dissertations you are writing
South African information systems departments, at UCT, Pretoria, Johannesburg, the Witwatersrand, KwaZulu-Natal, North-West and Unisa, examine two kinds of master’s dissertation, and the chapter structure differs between them.
The behavioural dissertation studies how people and organisations use technology: the adoption of mobile banking among small retailers, the factors behind e-learning continuance at a university, the readiness of a municipality for a new system. It borrows its theories from the field’s canon, the technology acceptance model, the unified theory of acceptance and use of technology, diffusion of innovations, and it usually ends in a survey analysed with regression or PLS-SEM, or in interviews analysed thematically.
The design science dissertation builds something: a framework, a model, a method or a working system, and evaluates it. It follows Hevner, March, Park and Ram’s seven guidelines from MIS Quarterly in 2004, which require a viable artefact, a relevant problem, rigorous evaluation and a clear research contribution, and it is usually organised around the six activities of Peffers, Tuunanen, Rothenberger and Chatterjee’s design science research methodology from the Journal of Management Information Systems in 2007.
The commonest structural failure in a South African IS dissertation is a design science study forced into the behavioural chapter order, so that the artefact appears in a chapter called Results and the evaluation is squeezed into the Discussion. Examiners circle it every time. Decide the tradition before the outline, and if you are unsure, the proposal is where to settle it; our guide to writing a research proposal for a South African university shows where that decision is recorded.
The behavioural structure, chapter by chapter
| Chapter | What it must contain | Field convention |
|---|---|---|
| 1 Introduction | Background, problem statement, research questions, objectives, scope, outline of chapters | The problem is an organisational or societal one, not a technical one |
| 2 Literature review and theoretical framework | The domain literature, then the theory that frames the study, ending in a research model with hypotheses | The research model figure closes the chapter; every hypothesis maps to an arrow |
| 3 Research methodology | Philosophy, approach, design, sampling, instrument, data collection, analysis, ethics | Instrument items are adapted from published studies and the source of each item is tabled |
| 4 Results | Sample profile, measurement model, structural model or thematic findings | Reliability and validity are reported before any hypothesis is tested |
| 5 Discussion | Each finding against the literature, with explanation for what did not hold | Unsupported hypotheses get as much space as supported ones |
| 6 Conclusion | Answers to the research questions, contributions, limitations, recommendations, future research | Contributions are split into theoretical, practical and methodological |
Chapter 2 is where behavioural dissertations run longest and weakest. The fix is to separate the domain review from the theory: review what is known about, say, mobile banking adoption among informal traders, then present the theory you will use, then combine them into your model. How to build the review itself, with the synthesis structure South African examiners expect, is in our guide to writing a literature review for a South African dissertation.
Chapter 4 has a field-specific requirement: if you used PLS-SEM, and most behavioural IS master’s dissertations now do, the measurement model must be reported in full before the structural model. The reporting conventions, and when PLS-SEM is the wrong choice for your sample, are worked through in our comparison of regression, SEM and PLS-SEM for a survey-based dissertation.
The design science structure, chapter by chapter
| Chapter | What it must contain | Maps to |
|---|---|---|
| 1 Introduction | The problem, why it matters, the research questions, the intended artefact and its scope | Peffers activity 1: problem identification and motivation |
| 2 Literature review | The problem domain, existing artefacts and their gaps, the kernel theories that inform the design | Grounds Hevner’s rigour guideline |
| 3 Research methodology | The design science paradigm, the process model followed, the evaluation strategy, ethics for user studies | Peffers activities as the chapter spine |
| 4 Requirements and design objectives | Objectives of a solution derived from the literature and from stakeholders, stated so that they can be evaluated | Peffers activity 2: objectives of a solution |
| 5 Design and development | The artefact: architecture, components, design decisions and the alternatives rejected | Peffers activity 3 |
| 6 Demonstration and evaluation | The artefact applied to a case or with users, then evaluated against the objectives of chapter 4 | Peffers activities 4 and 5 |
| 7 Conclusion | Contributions as artefact, design principles and knowledge; limitations; future work | Peffers activity 6: communication |
Two conventions distinguish a strong design science dissertation. First, the evaluation must be planned in chapter 3 and executed in chapter 6 against the objectives written in chapter 4; an evaluation invented after the artefact was built reads as an afterthought. Second, the contribution in chapter 7 is not the software. It is what the design taught: the design principles, the model, the method that someone else could apply. A working system with no articulated principles is a project, not a dissertation, and the external examiner will say so.
The user studies that most evaluations require need ethics clearance like any other human-participant research, and if your artefact processes personal information the Protection of Personal Information Act applies to the test data as well as the production data. What the committee will ask is set out in our explainer on ethics clearance at a South African university.
Where South African IS dissertations lose the year
Building before writing. Design science students spend eight months on the artefact and four weeks on six chapters. Write chapters 1, 2 and 4 before the first line of code; they change less than you fear, and they define what you are building.
A model with no hypotheses. Behavioural students draw the research model in the proposal and forget that each arrow is a hypothesis to be stated, numbered and tested. If chapter 2 ends without a numbered list of hypotheses, chapter 4 has nothing to report against.
A literature review that is a product review. Existing systems belong in the review, but as evidence of a gap, not as a catalogue. Each artefact you describe should end with what it does not do and why that matters for your problem.
Publishing late. South African IS departments expect a conference paper from a master’s, typically at the annual SAICSIT conference, or a journal article in the South African Computer Journal or the South African Journal of Information Management. Draft the paper from chapters 4 to 6 as soon as they exist; the reviewers’ comments improve the dissertation, and a submitted paper strengthens the case for a bursary renewal.
A discussion that repeats the results. Both traditions need a discussion that argues rather than restates; what that argument looks like is in our guide to writing the discussion chapter of a dissertation.
What the external examiner’s report asks, chapter by chapter
South African examiner report templates ask a fixed set of questions, and each maps to a chapter. Is the research problem clearly stated and its significance argued? That is chapter 1, and an examiner who cannot restate your problem in one sentence after reading it will say so on the first page. Does the candidate demonstrate command of the literature and identify a genuine gap? Chapter 2, and the gap must be stated in words, not implied by the absence of a reference. Is the methodology appropriate to the questions and adequately justified? Chapter 3, and for design science the word the examiner looks for is evaluation: what was measured, against what, and why that counts as evidence. Are the findings valid, and are the conclusions supported by them? Chapters 4 to 6, and the commonest note here is a conclusion that claims more than the evaluation or the sample can carry. Is the contribution to knowledge clear and original at master’s level? The final chapter, and it must name the contribution rather than list what was done. Write each chapter with its question beside you and the report writes itself in your favour.
The two references every IS dissertation cites, in Harvard
Both traditions cite the same two papers, so get them right once. Reference list: Hevner, A.R., March, S.T., Park, J. and Ram, S. (2004) ‘Design science in information systems research’, MIS Quarterly, 28(1), pp. 75–105. And: Peffers, K., Tuunanen, T., Rothenberger, M.A. and Chatterjee, S. (2007) ‘A design science research methodology for information systems research’, Journal of Management Information Systems, 24(3), pp. 45–77. In text, four authors are usually cited in full at first mention and as Hevner et al. (2004) thereafter, but UCT, Pretoria, Wits, Stellenbosch and Unisa each publish their own Harvard variant, so check whether your faculty wants single quotation marks around the article title and how it abbreviates page ranges.
Getting from outline to draft in Tesify
The structure above is the part Tesify does first. You choose the tradition, behavioural or design science, and it builds the chapter outline in that order, with the subsections each chapter needs, so that the supervisor’s first comment is about your argument rather than your headings. From there it drafts each section from your own material: your research model and hypotheses, your artefact description and design decisions, your evaluation results. The reference list is held in your faculty’s Harvard variant and rebuilt whenever a citation changes, and the AI editor corrects the register of a chapter written at midnight after a day at work. The free plan covers the outline and the first drafts.
The work stays yours. The system drafts from what you give it, the artefact and the data are your own, and the similarity check runs before your faculty runs Turnitin. Whether that is permitted at your institution depends on its policy, and we have read the policies; the summary is in our answer to whether AI is allowed for dissertations at South African universities.
Start your information systems dissertation outline, free
Frequently asked questions
How many chapters does an information systems master’s dissertation have?
Six in the behavioural tradition and usually seven in design science, where requirements and evaluation each need their own chapter. Departments differ, so confirm the template with your supervisor before the proposal.
Can I combine a survey with an artefact?
Yes, and many South African IS dissertations do, using a survey or interviews to derive requirements in chapter 4 and a user evaluation in chapter 6. The dissertation is still a design science study and should be structured as one.
Do I need the artefact to work fully?
No. A prototype that demonstrates the design against the stated objectives is sufficient at master’s level. What must be complete is the evaluation of what you built, honestly reported.
What does Tesify cost?
The free plan covers the chapter outline and first drafts. Paid plans are listed on the pricing page; there is no contract and you can stop after the chapter you needed help with.
Will my university treat this as plagiarism?
Not if you use it as a drafting and editing tool on your own material and follow your institution’s disclosure rules, which several South African universities now publish. Submitting generated text as your own research is misconduct under every policy, and Tesify is built for the former use, not the latter.
Is my prototype code or evaluation data safe?
Your dissertation content is stored under your account and is not used to train models or shared with other users. For personal information collected during evaluation, POPIA applies to you as the researcher, so anonymise before uploading.