How to Write a Research Findings Chapter Without Just Dumping Your Results

You have completed your analysis.

Your statistical software has produced tables, coefficients, p-values, confidence intervals, and charts.

Or your qualitative analysis has produced codes, themes, subthemes, and pages of participant quotations.

Now you need to turn all of that into a findings chapter.

This is where a common problem appears:

Students confuse presenting findings with presenting everything the analysis produced.

A findings chapter is not a storage room for SPSS output, every questionnaire response, every participant quotation, or every code generated during qualitative analysis.

Its job is to present the evidence needed to answer your research questions or objectives in a clear and logically organised way.

A useful principle is:

Research question → relevant analysis → key finding → supporting evidence

The reader should be able to move through the chapter and understand what the study found without having to reconstruct the analysis themselves.

Findings, results, or results and discussion?

Different universities and disciplines use different chapter structures.

You may encounter:

• Results

• Findings

• Results and Findings

• Results and Discussion

• separate Findings and Discussion chapters; or

• another discipline-specific structure.

There is no universal chapter title that every dissertation must use.

Quantitative studies often use the term results, while findings is common in qualitative research, but institutional conventions vary.

More importantly, some programmes expect results to be presented separately from interpretation against previous literature, while others combine findings and discussion.

Follow:

• your university guidelines;

• departmental requirements;

• supervisor guidance; and

• disciplinary conventions.

This article uses findings chapter broadly to mean the section where the evidence produced by your analysis is presented.

Know what belongs in the chapter

Before writing, distinguish between:

• raw data,
• analysis, and
• reported findings.

Suppose 350 students completed your questionnaire.

The raw data might contain hundreds of rows and dozens of variables.

Your analysis might generate:

• frequencies;

• descriptive statistics;

• correlations;

• regression models;

• diagnostic information;

• charts; and

• supplementary analyses.

The findings chapter does not automatically need all of them.

It needs the results that help answer the study's questions and allow the reader to evaluate the evidence appropriately.

The same principle applies to qualitative research.

Twenty interview transcripts may generate hundreds of initial codes.

Your findings chapter should not simply list those codes.

It should present the developed analytical themes and the evidence supporting them.

Return to your research questions and objectives

The easiest way to lose control of a findings chapter is to organise it around whatever your software produced first.

Instead, return to the structure of the study.

Suppose your objectives are:

1. To describe undergraduate students' levels of academic self-efficacy.

2. To compare academic self-efficacy between nursing and engineering students.

3. To examine the relationship between academic self-efficacy and academic performance.

Your findings chapter could follow those objectives.

For example:

4.1 Participant characteristics

Provide relevant information about the analysed sample.

4.2 Academic self-efficacy

Present evidence addressing Objective 1.

4.3 Differences between nursing and engineering students

Present evidence addressing Objective 2.

4.4 Relationship between self-efficacy and academic performance

Present evidence addressing Objective 3.

This gives the chapter an analytical purpose.

Your research aims and objectives become part of the chapter's architecture rather than something mentioned only in Chapter One.

Do not force one section per objective when it makes no sense

Organising around objectives is useful, but it is not a rigid rule.

Sometimes:

• two objectives are addressed by the same analysis;

• one objective requires several substantial subsections;

• qualitative themes cut across several objectives; or

• mixed-methods integration requires another structure.

The goal is logical alignment, not mechanical symmetry.

Ask:

Can the reader see how each major section contributes to answering the study?

If yes, the structure is probably working.

Open the chapter briefly

A findings chapter usually does not need another long introduction to the entire dissertation.

A concise opening can tell the reader:

• what the chapter presents;

• how the findings are organised; and

• any information necessary to understand the structure.

For example:

This chapter presents the findings of the study in relation to the three research objectives. It begins with a description of the analysed sample, followed by findings concerning academic self-efficacy, programme differences, and the relationship between self-efficacy and academic performance.

That is enough to orient the reader.

Do not spend several pages repeating:

• the research background;

• the literature review;

• the full methodology; or

• definitions already established elsewhere

unless your institutional structure genuinely requires it.

Report the analysed sample accurately

If 400 participants were recruited but only 347 cases were included in a particular analysis, do not casually describe every result as:

N = 400

The findings chapter should make clear what data actually contributed to the reported analyses.

Depending on the study, you may need to report:

• number recruited;

• number who responded;

• exclusions;

• missing data;

• final analytical sample; and

• different sample sizes across analyses.

Article #32's quantitative-data-analysis guide emphasised checking the actual N used in an analysis rather than assuming it equals the number originally recruited.

That distinction should remain visible when you write the findings.

Describe participant characteristics selectively

Readers often need enough information to understand the sample.

Relevant characteristics might include:

• age;

• programme;

• year of study;

• employment status;

• study location; or

• another variable important to the research design.

But do not automatically report every demographic variable collected.

If participants were asked their preferred social-media platform but that information has no relevance to the study or interpretation of the sample, it may not deserve space in the main findings chapter.

Report characteristics that help readers understand who contributed the evidence.

Quantitative findings should answer analytical questions

A quantitative findings chapter should not read like this:

• Mean = 3.72.
• SD = 0.61.
• t = 2.34.
• p = .021.
• r = .38.
• R² = .24.

Those numbers may all be correct.

But the reader should not have to infer what they mean.

A stronger presentation connects the statistic to the question.

For example:

Nursing students had a higher mean self-efficacy score than engineering students. The estimated difference was 0.31 points, with the inferential analysis indicating evidence of a difference between the groups under the specified model.

The exact statistics required should then be reported according to your discipline's conventions.

The prose tells the reader what was found.

The statistics provide the supporting evidence.

Present descriptive results before relevant inferential results

Suppose you are comparing stress scores between two groups.

It is usually helpful to show readers what the groups actually look like before presenting the inferential test.

For example:

• Group A: mean, standard deviation, sample size.

• Group B: mean, standard deviation, sample size.

Then present the relevant comparison.

This helps readers interpret the magnitude and direction of the difference rather than seeing only:

p = .014

Descriptive and inferential statistics perform different jobs.

Your findings chapter should allow them to work together.

Do not report only p-values

A result such as:

p = .003

does not tell the reader:

• what was compared;

• which direction the difference went;

• how large it was;

• how precise the estimate was; or

• whether it matters substantively.

Where appropriate, report information such as:

• group estimates;

• differences;

• correlation coefficients;

• regression coefficients;

• effect measures;

• confidence intervals;

• sample sizes; and

• relevant test statistics.

The exact combination depends on the analysis and reporting conventions.

But the findings should communicate magnitude and uncertainty, not simply whether a conventional significance threshold was crossed.

Report non-significant findings too

Suppose your study had three planned hypotheses.

Two produced statistically significant results.

One did not.

Do not quietly omit the third because it looks disappointing.

If an analysis was part of the study's planned attempt to answer a research question or test a hypothesis, its result is part of the findings.

Selective reporting can give readers a distorted picture of the evidence.

A non-significant result can still be informative, particularly when you report:

• the estimated effect;

• confidence interval;

• sample size; and

• relevant context.

Do not turn the findings chapter into a collection of only the results you hoped to obtain.

Unexpected findings belong if they are relevant

Your analysis may reveal something you did not predict.

For example, perhaps study time has little association with performance overall, but the data reveal a strong difference between full-time and part-time students.

Whether that belongs in the main findings depends on:

• its relevance;

• whether the analysis was planned or exploratory;

• the study's scope; and

• your reporting framework.

If you include an exploratory result, label it honestly.

Do not rewrite the history of the study to make an unexpected analysis look pre-planned.

Tables should reduce repetition

A table is useful when it presents information more efficiently than prose.

Suppose you need to report descriptive statistics for eight variables.

Writing eight paragraphs containing means, standard deviations, minima, and maxima may be tedious.

A clear table could communicate the information more efficiently.

Then the prose can highlight what matters.

Do not write:

Table 4.2 shows that Variable A had a mean of 3.4 and standard deviation of 0.7. Variable B had a mean of 4.1 and standard deviation of 0.6...

if the reader can already see every number in the table.

Instead, use the prose to identify the important pattern.

For example:

As shown in Table 4.2, perceived lecturer support had the highest average score among the measured support dimensions, while institutional belonging showed greater variability across participants.

The table carries the detail.

The prose provides direction.

Do not duplicate the same information three times

A common findings problem is:

• present a table;

• repeat every value in prose; and

• present a chart showing the same values again.

That is rarely necessary.

Ask what each format contributes.

Use:

• tables when precise values matter;

• figures when patterns or comparisons are easier to see visually; and

• prose to guide interpretation of the important evidence.

If all three communicate exactly the same thing, simplify.

Give every table and figure a purpose

Do not include a chart merely because your software generated one.

Every table or figure should:

• contribute to a research question or objective;

• communicate information clearly;

• have an informative title or caption;

• be referenced in the text; and

• be understandable without excessive detective work.

A reader should know why they are looking at it.

Do not paste statistical-software output into the chapter

Screenshots from SPSS, R output, Excel, or another statistical package are generally poor substitutes for properly prepared results tables.

Software output may contain:

• irrelevant statistics;

• internal labels;

• excessive decimals;

• confusing abbreviations;

• diagnostics not needed in the main text; and

• formatting inconsistent with your dissertation.

Create clean tables containing the information the reader actually needs.

Your software performs calculations.

It does not design your findings chapter.

Use sensible numerical precision

Do not report:

Mean = 3.726483917

when the measurement and research context do not justify that level of precision.

Use a consistent number of decimal places appropriate to:

• the measure;

• statistical convention;

• institutional style; and

• interpretation.

Excessive decimals create an illusion of precision.

Too little precision can also hide useful information.

Be consistent and purposeful.

Qualitative findings require a different presentation logic

A qualitative findings chapter is not simply the quantitative structure with quotations instead of numbers.

If you conducted thematic analysis, the chapter will often be organised around the themes developed through analysis.

Suppose Article #31's worked example produced a theme:

Performing competence to protect belonging

A findings section might:

• introduce the theme;

• explain its central meaning;

• develop the analytical pattern;

• provide carefully selected participant extracts;

• interpret those extracts; and

• show relevant variation or contradiction.

The theme becomes an analytical argument supported by participant evidence.

It should not become a container into which vaguely related quotations are dropped.

Introduce a qualitative theme before quoting participants

Avoid beginning a section with a quotation and leaving the reader to guess why it matters.

First establish the analytical point.

For example:

Participants frequently described academic uncertainty as something that needed to be concealed. Appearing competent became connected to whether they felt entitled to belong in university spaces.

Then provide an appropriate participant extract.

After the quotation, continue the analysis.

The structure becomes:

Analytical claim → evidence → interpretation

rather than:

Quotation → quotation → quotation

Article #31's thematic-analysis guide explains why quotations support analysis rather than replace it.

Use quotations selectively

A powerful quotation can make a qualitative finding vivid.

Twenty similar quotations can make the section exhausting.

Choose extracts because they:

• illustrate the analytical point;

• show important variation;

• reveal complexity;

• provide particularly clear evidence; or

• challenge a simple interpretation.

Do not select quotations only because they sound dramatic.

And do not repeatedly quote the same participant simply because they were especially articulate.

Your evidence should reflect the analytical pattern you are claiming.

Give quotations enough context

Consider:

“I just stopped going.”

Without context, the reader may not know:

• what the participant stopped attending;

• what happened beforehand;

• why the statement matters; or

• how it connects to the theme.

Sometimes a slightly longer extract is needed.

But avoid reproducing huge transcript sections when only a few sentences contribute to the point.

Use enough context to preserve meaning.

Protect participant identity

Quotations may contain identifying details even after names are removed.

For example, a participant might say:

“As the only female neurosurgeon at Hospital X...”

Replacing the participant's name with Participant 4 may not make that quotation anonymous.

Before reporting extracts, consider:

• occupation;

• location;

• rare experiences;

• job titles;

• organisations;

• family relationships; and

• combinations of characteristics

that could make someone identifiable.

Follow your consent process, ethics approval, institutional requirements, and data-protection procedures.

Show variation instead of pretending everyone agreed

A theme does not require every participant to have the same experience.

Suppose a theme concerns lecturer support.

Most participants may describe lecturer recognition as strengthening belonging.

Several others may describe lecturer attention as uncomfortable or patronising.

That variation may deepen the theme.

You might write:

Although lecturer recognition generally strengthened participants' sense of belonging, this was not universal. For some participants, highly visible support reinforced their sense of being different from classmates.

Then support both aspects with appropriate evidence.

Qualitative findings become stronger when they acknowledge meaningful complexity.

Do not organise qualitative findings participant by participant

A weak qualitative findings chapter sometimes becomes:

Participant 1 said...

Participant 2 said...

Participant 3 said...

This may summarise the interviews, but it does not necessarily present the patterns developed through analysis.

If your analytical approach produced themes, organise the findings primarily around those themes rather than retelling each interview separately.

Individual participants then provide evidence within the analytical structure.

For example:

Theme 1: Performing competence to protect belonging

Explain the pattern and support it with extracts from relevant participants.

Theme 2: Recognition made belonging feel legitimate

Develop the second pattern using evidence across the dataset.

This allows the chapter to show what the analysis found across participants, while still preserving important individual variation.

Subthemes should add useful structure

A complex theme may contain several related subthemes.

For example:

Theme: Performing competence to protect belonging

Possible subthemes:

hiding uncertainty from peers;

reluctance to seek academic help; and

achievement as evidence of legitimacy.

Subthemes can make a complicated analytical pattern easier to communicate.

But do not create them simply because a dissertation looks more sophisticated with additional headings.

Each subtheme should contribute something distinct to the broader theme.

Findings and discussion are not always separated

Students are sometimes told:

“Never interpret anything in the findings chapter.”

That advice is too absolute.

The appropriate boundary depends on your discipline, methodology, and institutional structure.

In a traditional quantitative dissertation with separate Results and Discussion chapters, the Results chapter may focus heavily on presenting the statistical evidence, while broader explanation and comparison with previous literature are reserved for Discussion.

In qualitative research, presentation and interpretation may be harder to separate because explaining what a theme means is itself part of the analysis.

Some dissertations therefore combine findings and discussion.

The correct question is not:

“Am I allowed to interpret?”

It is:

“What kind of interpretation belongs here under the structure required for my study?”

Distinguish analytical interpretation from broader discussion

Even when Findings and Discussion are separate, qualitative findings usually need some analytical explanation.

Suppose a participant says:

“I never asked questions because everyone else looked like they understood.”

Simply reproducing that quotation is not analysis.

You may explain that the extract illustrates how perceived peer competence made uncertainty feel risky to reveal.

That is interpretation of the data.

A broader Discussion chapter might then connect that pattern to:

previous studies;

theory;

the conceptual framework;

alternative explanations; and

implications.

The exact boundary varies, but the distinction is useful:

Findings: What pattern does the evidence show?

Discussion: What does that finding mean in relation to the wider academic conversation?

Do not introduce large amounts of new literature accidentally

If your dissertation has a separate Discussion chapter, the Findings chapter usually should not become another literature review.

Avoid writing:

Participants reported reluctance to seek help. Smith (2022), Jones (2023), Ali (2024), and Chen (2025) also found...

after every result unless your required structure explicitly combines findings and discussion.

That literature comparison normally belongs in the Discussion.

Keeping the roles clear helps prevent repetition across chapters.

Mixed-methods findings need deliberate integration

A mixed-methods study may contain both quantitative and qualitative findings.

There is no single required structure.

Depending on the design, you might present:

• quantitative findings first, followed by qualitative findings;

• qualitative findings first, followed by quantitative findings;

• findings by research question with both forms of evidence integrated; or

• separate phases followed by an explicit integration section.

The structure should follow the mixed-methods design.

Do not simply write one quantitative chapter and one qualitative chapter and assume that placing them next to each other constitutes integration.

Look for convergence, complementarity, and divergence

Suppose your survey finds that students generally report high lecturer support.

Interviews may reveal that students value lecturer availability but differ in whether they feel comfortable using that support.

The two datasets are not necessarily contradictory.

The qualitative evidence may add depth to the quantitative pattern.

You might encounter:

• Convergence: Both forms of evidence point toward a similar conclusion.

• Complementarity: One dataset helps explain or elaborate the other.

• Divergence: The datasets appear to point in different directions.

Divergence should not automatically be treated as an analytical failure.

It may reveal differences in:

• measurement;

• context;

• participant interpretation;

• sampling;

• timing; or

• the phenomenon itself.

The important thing is to address the relationship between the findings rather than hiding inconvenient differences.

Do not force different datasets to agree

Imagine questionnaire responses suggest that students are satisfied with academic feedback, while interviews reveal repeated frustration about its usefulness.

Do not selectively report only the evidence that makes the two datasets appear consistent.

Instead, investigate the difference.

Perhaps the questionnaire measured general satisfaction while interviews explored whether feedback helped students improve future work.

Those may be related but distinct issues.

A mixed-methods study can become more informative when different forms of evidence reveal complexity.

Use headings that communicate findings

Compare:

4.3 Research Objective Two

with:

4.3 Students with stronger academic self-efficacy reported higher engagement

The first tells readers where they are in the dissertation.

The second tells them something about what was found.

Depending on your institutional style, you may combine both:

4.3 Objective Two: Relationship Between Academic Self-Efficacy and Engagement

For qualitative findings, analytical theme names can perform a similar role.

Headings should help readers follow the argument.

Build transitions between sections

A findings chapter should not feel like separate blocks pasted together.

At the end of one major section, you can briefly indicate what comes next.

For example:

Having described participants' levels of academic self-efficacy, the next section examines whether these scores differed between nursing and engineering students.

This creates continuity without repeating the previous section.

Transitions are particularly useful when the chapter moves between:

• descriptive and inferential findings;

• quantitative and qualitative evidence;

• different research objectives; or

• themes and subthemes.

Keep methodological explanation proportionate

Sometimes readers need a brief reminder of what an analysis represents.

But the findings chapter should not reproduce the entire Methods chapter.

Instead of writing several paragraphs re-explaining how Pearson correlation works, you might state the relevant analysis and report its results.

If an unexpected analytical decision needs explanation, provide enough information for readers to understand it and refer to the methodology where appropriate.

The chapter's primary job remains presenting findings.

Report negative and null findings honestly

A study does not become unsuccessful because the expected result did not appear.

Suppose you hypothesised that study time would be positively associated with academic performance, but the analysis did not provide convincing evidence of that relationship.

That is still a finding.

Report it accurately.

Likewise, qualitative participants may reject an assumption that shaped your initial research expectations.

Do not hide those accounts because they complicate the story.

Research findings should reflect the evidence, not the result you hoped to obtain.

Unexpected findings can be valuable

An unexpected result may become one of the most interesting parts of a study.

Suppose your research focused on lecturer support, but interviews repeatedly reveal that commuting determines whether students can participate in informal academic activities.

If that pattern is relevant to the research question, it may deserve reporting.

But maintain transparency.

If the analysis was exploratory, say so where appropriate.

Do not pretend the study was designed around a finding discovered afterward.

Do not overstate what the evidence supports

Compare:

Students who studied longer achieved higher scores.

with:

Longer reported study time was associated with higher assessment scores in the analysed sample.

The second statement is more careful when the design supports association but not causation.

Similarly, a qualitative study involving one institution should not automatically conclude:

All university students experience...

Match your claims to:

• the research design;

• sample;

• measurement;

• analysis; and

• uncertainty.

Strong academic writing does not require exaggerated certainty.

A worked quantitative findings example

Suppose your objective is:

To examine the relationship between weekly study time and academic performance among first-year undergraduate students.

Weak version

A Pearson correlation was performed. The correlation was .36 and the p-value was .002. Therefore, the hypothesis was accepted.

Several problems appear:

• the variables are not clearly described;

• sample size is missing;

• uncertainty is not communicated;

• “accepted” overstates what the statistical result establishes;

• the direction is reported but barely interpreted; and

• the finding is disconnected from the research question.

Stronger structure

A stronger presentation would:

1. briefly describe study time and academic-performance scores;

2. identify the analysis used;

3. report the correlation estimate;

4. report appropriate uncertainty and inferential information;

5. state the direction and magnitude in context; and

6. connect the result to the research question without claiming causation.

For example:

Weekly study time was positively associated with academic-performance scores in the analysed sample, indicating that students reporting more study time tended to report higher performance. The estimated correlation was moderate in magnitude.

The exact statistical details should accompany that statement according to the required reporting style.

The prose explains the finding.

The statistics document the evidence.

A worked qualitative findings example

Suppose your theme is:

Performing competence to protect belonging

Weak version

Participant 3 said, “I didn't ask because everyone looked like they understood.” Participant 7 said, “I didn't want people thinking I was stupid.” Participant 11 said, “I kept quiet.”

This provides quotations but little analysis.

Stronger structure

Begin with the analytical claim:

Participants often described concealing academic uncertainty because appearing competent was tied to whether they felt they legitimately belonged among their peers.

Then provide a carefully selected quotation.

After the quotation, explain what it contributes:

The extract illustrates how silence functioned as a form of self-protection. Asking for clarification risked making uncertainty visible in an environment where the participant perceived others as already competent.

You might then present another extract that adds variation or develops the pattern.

The quotations support the theme.

They do not constitute the theme by themselves.

A worked mixed-methods example

Suppose a questionnaire finds:

78% of respondents agreed that lecturers were available when academic help was needed.

Interviews, however, reveal that some students rarely approached lecturers because they feared appearing academically weak.

A weak mixed-methods conclusion might say:

Both findings show lecturer support was good.

That misses the complexity.

A stronger integration might explain:

Although most survey respondents perceived lecturers as available, interview accounts suggested that availability did not always translate into help-seeking. Some participants avoided approaching lecturers because requesting help risked exposing academic uncertainty.

Now the qualitative evidence explains an important distinction hidden by the survey percentage.

That is integration.

Common findings-chapter mistakes

Reporting everything the software produced

The findings chapter should contain evidence relevant to the study, not every available statistic.

Organising the chapter around software output

Research questions and analytical logic should determine the structure.

Reporting only significant results

Planned non-significant findings are still findings.

Repeating tables word for word

Use prose to highlight meaning rather than duplicate every cell.

Using tables and charts for the same information without purpose

Each format should contribute something useful.

Pasting software screenshots

Prepare clean, reader-focused tables and figures.

Using quotations without analysis

Participant extracts support qualitative claims; they do not replace them.

Using too many quotations

More quotations do not automatically create stronger qualitative evidence.

Organising qualitative findings participant by participant

Where the analysis developed cross-dataset themes, present those patterns rather than retelling interviews individually.

Hiding contradictory evidence

Variation can strengthen the credibility and depth of the analysis.

Introducing the Discussion too early

If Findings and Discussion are separate, avoid turning every result into a literature comparison.

Refusing all interpretation

Qualitative findings often require analytical explanation even when broader theoretical discussion occurs later.

Treating mixed-methods sections as automatically integrated

Separate quantitative and qualitative results need an explicit analytical relationship.

Making causal claims from associative evidence

Use language that matches the design.

Overclaiming generalisability

Your conclusions should not exceed what the sample and design can support.

Repeating the methodology chapter

Explain only what readers need to understand the reported findings.

A findings-chapter checklist

Before finalising the chapter, ask:

1. Is the chapter organised around the research questions, objectives, themes, or another defensible analytical structure?

2. Can readers see how each major section contributes to answering the study?

3. Have I clearly identified the analysed sample where necessary?

4. Have I reported relevant participant characteristics without dumping unnecessary demographics?

5. Do quantitative sections present descriptive context before relevant inferential findings?

6. Have I reported more than p-values where appropriate?

7. Are effect estimates and uncertainty communicated appropriately?

8. Have I included planned non-significant findings?

9. Are exploratory findings identified honestly?

10. Does every table or figure have a clear purpose?

11. Have I avoided repeating every table value in prose?

12. Have I avoided presenting the same information unnecessarily as a table, figure, and paragraph?

13. Are tables and figures labelled clearly and referenced in the text?

14. Have I removed irrelevant software output?

15. Are numerical values reported with sensible and consistent precision?

16. Are qualitative findings organised around analytical patterns rather than participant order where appropriate?

17. Does each qualitative theme have a clear analytical meaning?

18. Are participant quotations used selectively as evidence?

19. Have I provided enough context for quotations to make sense?

20. Have I considered confidentiality when reporting quotations?

21. Have I represented meaningful variation and contradiction?

22. If the study is mixed methods, have I actually integrated the different forms of evidence?

23. Have I kept Findings and Discussion appropriately separated or integrated according to my required structure?

24. Have I avoided causal claims the research design cannot support?

25. Have I avoided generalising beyond the evidence?

26. Does the chapter end with a clear picture of what the study found?

If several answers are no, the chapter probably needs another revision before you move to the Discussion or Conclusion.

How to end the findings chapter

The ending does not need to repeat every statistic and theme.

Instead, briefly remind the reader what the chapter accomplished and create a bridge to what follows.

For a dissertation with a separate Discussion chapter, you might explain that the next chapter interprets the findings in relation to:

• the research questions;

• previous literature;

• theoretical or conceptual framework;

• implications; and

• study limitations.

For a combined Findings and Discussion structure, the transition may instead lead toward synthesis or the final conclusion.

Keep the ending proportional.

A thirty-page findings chapter does not need another five-page summary of itself.

Final takeaway

A strong findings chapter does not prove how much analysis you performed.

It shows the reader what the analysis revealed.

Organise the chapter around the questions the study set out to answer.

Present enough information to make the evidence understandable.

Use statistics to support quantitative findings rather than overwhelm them.

Use themes and quotations to communicate qualitative patterns rather than reproduce transcripts.

Use tables and figures when they make evidence clearer.

Report results that did not match your expectations.

Preserve meaningful contradictions.

And match every claim to what the research design and analysis can genuinely support.

The most useful question while writing the chapter is therefore not:

“Have I included all my results?”

It is:

“Have I presented the evidence the reader needs to understand what my study actually found?”

When the answer is yes, the findings chapter becomes more than a collection of outputs.

It becomes a clear account of the evidence your research produced.