How to write up quantitative results for UK dissertations

Written By

John Miller

✔️ 97% Satisfaction | ⏰ 97% On Time | ⚡ 8+ Hour Delivery

Keyword: write quantitative results dissertation UK

Writing up quantitative results requires presenting numbers clearly. Readers need to understand findings without confusion. You're translating statistics into English. This translation is key.

There's a particular kind of stress that comes with knowing you've got a deadline and not feeling ready. It can make it harder to think clearly, harder to write, and harder to ask for help when you need it most. We get that. We've designed our service around making it as easy as possible to get the support you need quickly, without having to work through complex systems or wait days for a response.

The process of writing a methodology chapter teaches you far more about your chosen subject than you would learn from passive reading alone, because it forces you to engage with the material at a level of depth that other forms of study rarely demand from students at this stage of their academic careers.

Your results section reports findings. It doesn't interpret or discuss. That's the next section. Results sections are straightforward reporting. They're often the hardest to write because they're constrained by what data actually shows.

#### Organising Your Results Section

Start with descriptive statistics. Describe your sample. Report means and standard deviations (or frequencies and percentages for categorical data). Help readers understand what you measured.

Then report main analyses. If your research question asks whether groups differ, present group comparisons. If asking whether variables correlate, report correlations. Structure mirrors research questions.

Follow logical order. Simpler analyses before complex ones. Univariate (single variable) before multivariate (multiple variables). Readers follow easier content before tackling complex material.

#### Presenting Descriptive Statistics

The amount of time you spend on planning and outlining before you start writing will save you at least twice as much time during the drafting and revision stages, making it one of the wisest investments you can make.

Use tables. "Table 1: Descriptive Statistics for Study Variables" with rows for each variable and columns for N, M (mean), and SD (standard deviation).

Or present in text. "Participants (N=234) had mean age 42.3 years (SD=15.1) and included 58 per cent female respondents."

For group comparisons, separate by groups. "Control group (n=100): M=45.3, SD=8.2; Intervention group (n=134): M=52.1, SD=9.4."

Include sample sizes. Readers need context. N matters. They assess whether conclusions are justified given sample size.

#### Reporting Statistical Tests

Don't just state results. Include statistics that let readers verify. For t-tests: "t(232)=5.23, p<.001, d=0.73." That means t-statistic 5.23, df=232 (sample size calculation), p<0.001 (quite considerable), effect size d=0.73 (large).

Your appendices give you a place to include supporting material that strengthens your dissertation without interrupting the flow of your main argument, such as additional data, sample materials, or detailed calculations.

For ANOVA: "F(2,231)=8.45, p<.001, ω²=0.06." F-statistic, numerator and denominator df, p-value, effect size.

For correlation: "r(232)=.45, p<.001." Correlation coefficient and p-value.

For regression: Include R², F-statistic, coefficients with standard errors, t-values, p-values.

These details look intimidating. They're key. They let readers and reviewers assess your findings. Include them always.

#### Using Tables Effectively

Tables present results efficiently. Don't write paragraphs when tables work better.

"Table 2: Correlation Matrix of Study Variables" with variables in rows and columns, correlations filling cells. Include sample size and significance notes.

"Table 3: Regression Analysis Predicting Exam Performance" with predictors in rows, columns for coefficient, SE, t, p, and beta.

The experience of completing a dissertation prepares you for many of the challenges you will face in professional life, including managing complex projects, communicating clearly, and working independently towards a considerable goal.

Tables should be self-contained. Someone reading the table without surrounding text should understand it. Include clear headers, notes explaining abbreviations, significance indicators (, , ).

#### Reporting in Text

Write results clearly. "The intervention group scored higher than control group (t=5.23, p<.001), with large effect size (d=0.73)." Readers immediately understand.

Connect to research questions. "This supports hypothesis 1 that the intervention improves outcomes."

Report negative findings. "Contrary to prediction, gender didn't moderate intervention effects (F=0.32, p=.57)."

Include confidence intervals. "Intervention improved scores by average of 6.8 points (95 per cent CI = 3.2 to 10.4)." Readers see uncertainty. They understand precision.

#### Dealing with Unexpected Findings

Sometimes results surprise you. They don't support your hypotheses. That's fine. Report honestly. "Hypothesis 2 predicted X would predict Y. However, the relationship was non-considerable (r=.12, p=.18)."

Explore unexpected patterns. What might explain surprising results? Be cautious. You're theorising after seeing data. Call it exploratory. Don't overstate.

Include effect sizes even for non-considerable findings. "Although non-considerable (p=.18), the effect size was small (d=0.18), suggesting the relationship is likely trivial."

#### Checking Assumptions

Report assumption tests. Did your data meet statistical assumptions?

For t-tests and ANOVA: "Levene's test indicated homogeneity of variance (F=1.23, p=.29). Shapiro-Wilk test indicated normality (p=.67)." Results met assumptions.

Or: "Shapiro-Wilk test indicated non-normality (p<.05). However, with large sample (n=234), t-test is strong to violations."

Transparency matters. Show you've checked. Show your data justified your analyses.

#### Presenting Complex Results

Moderation analysis shows whether effects differ across conditions. "We tested whether gender moderated intervention effects. Moderation wasn't considerable (ΔR²=.01, p=.42)." Clear and concise.

Mediation analysis examines mechanisms. "The intervention's effect on outcomes was mediated by increased motivation (indirect effect=0.34, SE=0.11, 95 per cent CI=0.13-0.56)." Shows the pathway.

Multilevel analysis for nested data. "Controlling for school clustering (ICC=0.12), the intervention effect remained considerable (b=6.23, SE=1.45, p<.001)." Shows you've handled data structure.

#### Dissertation Homework Support

Dissertation Homework supports students writing clear results sections. Your findings deserve transparent reporting. Readers should understand what you found. Your supervisor reviews results sections. They check that you've reported appropriately. They suggest clarifications.

Universities like Imperial College London, University of Manchester, University of Cambridge, Durham University, and University of Edinburgh emphasise results clarity. Your institution provides writing support. Use it. Results sections often benefit from revision. Clarity improves with practice.

FAQ

Q1: Should I include raw data in my results section, or only summary statistics?

Include summary statistics, not raw data. Present means, standard deviations, group comparisons. Don't list every participant's score. That's overwhelming. Raw data belongs in appendices if needed. Summary statistics are efficient. They let readers understand patterns without information overload. Your supervisor confirms what summary information matters.

Q2: How should I report non-considerable findings?

Report them honestly. Include p-value and effect size. "The relationship was non-considerable (r=.12, p=.18) with small effect size (r=.12)." Show you measured it. Show whether non-significance reflects true absence of relationship or insufficient power. Non-considerable findings are still findings. Report them.

Q3: How many decimal places should I use when reporting statistics?

Two decimal places is standard. t(232)=5.23, not t(232)=5.2344. p=.001 (note leading zero), not p=.0009. Round appropriately. Don't suggest false precision. Your supervisor confirms discipline conventions. Consistency matters. All t-values have same decimal places. All p-values formatted consistently.

Q4: Should I interpret results in the results section, or save interpretation for the discussion?

Straight reporting in results. "Table X shows..." Minimal interpretation in results. Save discussion and explanation for the discussion section. Results sections are factual. What did you find? Avoid saying "This suggests..." or "This indicates..." That's discussion. Results sections present. Discussion sections interpret. This separation helps readers handle your dissertation clearly.

Q5: How do I report effect sizes when different test types have different effect size metrics?

Report appropriate effect sizes for tests used. T-tests: Cohen's d. ANOVA: eta-squared or omega-squared. Correlation: r. Regression: R². Always report effect sizes. They complement p-values. They show magnitude. Different metrics are fine. Readers understand context. Just report what's appropriate for your test.