Paul Scott
Strong dissertation recommendations connect a specific finding to a justified next step, while respecting what your research cannot establish. Each recommendation needs a clear audience, a purpose and a traceable link to the evidence.
The challenge is deciding whether your findings justify action, further investigation or a more cautious interpretation. Start by separating those possibilities before drafting your final chapter.
Key Takeaways
- Link every recommendation to a finding and explain the reasoning between them.
- Match the proposed action to your research design, limitations and intended audience.
- Distinguish practical recommendations from implications and suggestions for further research.
- Give future research a clear question and purpose, rather than requesting more research generally.
Should Your Thesis Include Recommendations?
Consider Who Can Act
Recommendations usually address someone who can make a decision. A research report might advise an organisation about its services, whereas a doctoral thesis primarily addresses examiners and a wider academic audience.
In Pat Thomson’s discussion of thesis recommendations, she argues that transferring report conventions into a thesis can create genre confusion. Your examiner can assess your contribution, but may have no authority to implement your proposals.
Thomson suggests conditional recommendations, framed as what you would recommend if making them. Another option is a separate leaflet, short paper or film for the audience able to act.
Separate Meaning From Proposed Action
An implication explains what the results mean for understanding a problem. A recommendation proposes what someone should do because of it.
Future-research recommendations identify an unresolved question and explain how another study could address it. Practical recommendations propose changes to policy, professional practice or organisational decisions.
These categories can overlap, but they need different support. Interviews may reveal how participants understand an experience without establishing how widespread it is. Likewise, an observed association doesn’t establish that changing one variable will cause another to improve.
Where Dissertation Recommendations Belong
Placement depends on disciplinary conventions and the required chapter structure. Recommendations may sit within the discussion, appear in the conclusion or occupy a separate section.
Scribbr’s dissertation structure guide describes this variation: some disciplines use a short conclusion before the discussion, whilst others use a final chapter containing recommendations for future research. Its checklist asks for relevant recommendations for research or practice and an explanation of the conclusion’s implications.
Whatever the arrangement, your conclusion should answer the main research question concisely and identify the research’s contribution. Recommendations shouldn’t displace that answer.
Our guide to writing a dissertation conclusion explains how to bring findings together without repeating every chapter. Develop the reasoning in the discussion first, then retain the recommendation and its essential justification in the closing chapter.
A separate recommendations section needs the same connection to the evidence as one embedded elsewhere.
Move From Findings to Defensible Recommendations
Identify the Finding and Its Boundary
Start with a finding that answers your research question. Then state exactly what the results establish, using the same terminology as your findings chapter.
Preserve distinctions such as attendance versus participation, or reported confidence versus measured performance. Otherwise, your recommendation may address something you didn’t investigate.
Next, identify the relevant boundary shaped by your methodology. Recruitment at one institution limits claims about other settings. Self-reported behaviour raises different concerns from recorded behaviour. Our guidance on explaining dissertation limitations connects these constraints to the conclusions they affect.
A limitation should change the scope of your recommendation, rather than appear as a disclaimer after an overstatement.
Explain the Action and Its Purpose
Identify who could act, what they could do and why that action follows. Include a setting or evaluation measure when the evidence supports one.
Use verbs that match the proposed commitment. “Explore” proposes investigation; “pilot” proposes a limited test; “introduce” proposes implementation. They aren’t interchangeable.
For future research, specify the uncertainty. If interviews identify a recurring experience, a survey could examine its prevalence using clearly defined measures. If your evidence captures one moment, longitudinal research could investigate change.
The reasoning between the finding and proposed action belongs in your dissertation discussion chapter, where you can compare explanations against your literature review.
Worked Example: Sixteen School-Leader Interviews
Identify the Unsupported Leap
Thomson uses interviews with sixteen school leaders to illustrate the problem of scale. Such interviews aren’t a sound basis for sweeping education-policy recommendations or professional development programmes.
However, they can deepen understanding of identities, events and meaning-making. Their contribution can therefore be substantial without establishing that a particular intervention works.
A recommendation for nationwide training would require additional reasoning and evidence about effectiveness, implementation and different settings. Increasing the confidence of the wording wouldn’t supply that evidence.
The defensible starting point is the understanding developed through the interviews, with its contextual limits intact.
Write a Bounded Recommendation
Using Thomson’s example, conditional wording could read:
If recommendations for professional development were required, I would recommend that providers explore the issues raised in these interviews with school leaders in their own settings. Any resulting programme would need evaluation before wider adoption.
This proposes investigation before implementation. It also names an audience with responsibility for professional development, rather than directing an examiner to change education policy.
In your dissertation, replace “the issues raised” with the precise theme your analysis established. Then identify the supporting evidence and explain why the proposed investigation would help.
For a humanities project, the corresponding recommendation may concern further interpretation or comparison. It doesn’t need to propose an organisational intervention to make a worthwhile contribution.
Formulate Future Research With EPICOT+
The BMJ’s 2006 research-recommendations editorial argues that simply calling for further research leaves the required work unspecified. It proposes EPICOT+ to refine research recommendations from systematic reviews and clinical guidelines about treatment effects.
The proposal followed a 2005 meeting involving the BMJ Publishing Group, Centre for Reviews and Dissemination, NICE, NCCHTA, SIGN and UK Cochrane Centre.
For an appropriate treatment question, address these elements.
| Element | What to Specify |
|---|---|
| Evidence | What existing evidence establishes and leaves uncertain |
| Population | Who the proposed research should study |
| Intervention | The treatment or intervention to examine |
| Comparison | What it should be compared with |
| Outcome | Which effects should be measured |
| Time Stamp | When the supporting evidence was assessed |
EPICOT+ builds on the standard PICO structure (Population, Intervention, Comparison, Outcome) by adding Evidence and Time Stamp.
The optional “+” elements concern disease burden, timeliness and study type. The time stamp dates the evidence base; it isn’t the proposed follow-up period.
Don’t force a literary, historical or interview-based project into an intervention framework. The editorial also cautions that the format may not transfer to diagnosis, prognosis or patient-preference research.
Outside treatment research, retain the underlying precision: identify the uncertainty, the material or population needed, and the analytical approach that could answer it. Explain what that answer would change about your current interpretation.
Edit for Scope, Feasibility and Consistency
Read each recommendation alongside its supporting finding. The action should address the same issue, population and setting, unless you’ve justified extending it.
Then consider feasibility. A proposal involving training, staffing or new data collection creates practical demands. Acknowledge those demands without inventing budgets, implementation schedules or likely benefits.
Prioritise the recommendations that matter most to your research question. An unrelated suggestion doesn’t become relevant because it’s easy to implement.
For a paper-based thesis, derive recommendations from the overarching argument. Explain where the papers agree, where they differ and how their methods affect the comparison. Simply collecting each paper’s recommendations won’t establish a thesis-wide position.
Finally, remove new results. Recommendations can synthesise evidence already analysed, but shouldn’t introduce findings that readers haven’t encountered.
Frequently Asked Questions About Dissertation Recommendations
How Many Recommendations Should I Include?
There isn’t a universal number. Include only recommendations with a distinct purpose and an identifiable evidence base.
Several versions of the same proposal usually belong together. Conversely, recommendations addressing different uncertainties need separate explanations. The required section length should reflect the dissertation’s scope, rather than encourage a longer list.
Is a Dissertation the Same as a Thesis?
Terminology varies. Scribbr explains that a dissertation generally means PhD research in the US, whilst in the UK it often means a bachelor’s or master’s project.
Its US-oriented account places committee selection and the prospectus before research and writing. It also says dissertations can take up to five years. That describes a different degree context from many UK taught projects; it doesn’t determine how extensive their recommendations should be.
Should Recommendations Appear in the Abstract?
Include a central recommendation only when it helps convey the study’s main conclusion. An abstract isn’t the place for a detailed action plan.
Scribbr’s abstract guidance gives a usual length of 150 to 300 words, covering the topic, aims, methods, main results and conclusions. Prioritise those elements, and retain any qualification needed to prevent an unsupported interpretation.
Keep the Next Step Within the Evidence
Strong dissertation recommendations make the connection between evidence and proposed action easy to follow. Their value comes from proportionate reasoning, clear responsibilities and an explicit purpose.
Before submitting, trace each recommendation back to a finding and the interpretation supporting it. If that connection is missing, narrow the proposal or remove it. A carefully bounded next step is easier to defend than an ambitious claim your study cannot support.