Why English? | Writing a Research Results Section Through Participant Flow, Descriptive Data, Primary Outcomes, Estimates, Uncertainty, Tables, Figures and Prespecified Analyses

ORIENTATION

What this guide covers

Current fact check

Checked on 9 November 2026, EQUATOR listed STROBE for cohort, case-control and cross-sectional observational studies and CONSORT 2025 for randomised trials. Current STROBE results items include participant numbers and reasons for non-participation, descriptive data and missingness, outcome events or summary measures, unadjusted and adjusted estimates with precision, category boundaries and other analyses. These are reporting expectations for applicable designs, not a universal analysis recipe.

What this guide does not cover

This article covers organising, drafting and checking a results section after authorised analysis. It does not cover study design, ethics approval, data collection, data cleaning decisions, statistical consultation, model selection, causal identification, fabrication, methods writing, discussion, peer review, journal submission or claims beyond the approved evidence.

Worked example

Every chapter returns to a fictional cross-sectional anonymous survey of study-planning habits among volunteer polytechnic students with a prespecified descriptive comparison and acknowledged voluntary-response limits. It is fictional, low-stakes and educational; it does not report a real school, clinical decision or guaranteed outcome.

Warm up with eduKate’s guides to explaining scientific evidence and asking better questions in class.

CHAPTER 1 OF 12

1. Return to the Question, Protocol and Analysis Plan

Back to contents

Why this move matters

List the outcomes, time points, groups and analysis order that were specified before inspecting the results.

Report what was found in the order the study asked

List the outcomes, time points, groups and analysis order that were specified before inspecting the results. For The survey’s primary result is weekly planning frequency, followed by a prespecified comparison by study load., begin with participant or material flow, descriptive data and the prespecified primary result before secondary or exploratory findings.

Pair every estimate with enough context to interpret it

Name groups, denominators, units, time points, effect estimates and uncertainty, and distinguish unadjusted from adjusted analysis. Preserve this limit: A manuscript should not rewrite the original plan to make an attractive finding primary.

Use tables and figures to carry evidence without duplication

Give each display a self-contained title, labels, notes and denominators, then use prose for the main pattern rather than reading every cell aloud.

Separate results from explanation and keep deviations visible

Report missing data, exclusions, harms, sensitivity analyses and departures from plan before interpretation. A manuscript should not rewrite the original plan to make an attractive finding primary. Clear reporting supports appraisal; it does not make an estimate causal or important by itself.

Student, parent and career transfer

Practise “Return to the Question, Protocol and Analysis Plan” by separating evidence, interpretation, decision and next action. A parent, teacher, supervisor or teammate can ask what supports the sentence and which missing fact could change it. This discipline transfers to school investigations, university research, laboratory reports, social science, health research, evaluation, data analysis, dissertations, evidence synthesis and research communication without guaranteeing grades, approval, publication, system safety, research validity, employment or future outcomes.

CHAPTER 2 OF 12

2. Report Participant or Material Flow

Back to contents

Why this move matters

State numbers invited, screened, included, excluded, completed and analysed with reasons at each relevant stage.

Report what was found in the order the study asked

State numbers invited, screened, included, excluded, completed and analysed with reasons at each relevant stage. For The section reports invitations, starts, completed surveys, duplicates removed and the final analysis count., begin with participant or material flow, descriptive data and the prespecified primary result before secondary or exploratory findings.

Pair every estimate with enough context to interpret it

Name groups, denominators, units, time points, effect estimates and uncertainty, and distinguish unadjusted from adjusted analysis. Preserve this limit: Flow numbers must reconcile across text, table and diagram.

Use tables and figures to carry evidence without duplication

Give each display a self-contained title, labels, notes and denominators, then use prose for the main pattern rather than reading every cell aloud.

Separate results from explanation and keep deviations visible

Report missing data, exclusions, harms, sensitivity analyses and departures from plan before interpretation. Flow numbers must reconcile across text, table and diagram. Clear reporting supports appraisal; it does not make an estimate causal or important by itself.

Student, parent and career transfer

Practise “Report Participant or Material Flow” by separating evidence, interpretation, decision and next action. A parent, teacher, supervisor or teammate can ask what supports the sentence and which missing fact could change it. This discipline transfers to school investigations, university research, laboratory reports, social science, health research, evaluation, data analysis, dissertations, evidence synthesis and research communication without guaranteeing grades, approval, publication, system safety, research validity, employment or future outcomes.

CHAPTER 3 OF 12

3. Describe Missing Data and Analysis Populations

Back to contents

Why this move matters

Show missingness by important variable and identify which records entered each analysis.

Report what was found in the order the study asked

Show missingness by important variable and identify which records entered each analysis. For The planning-frequency item has three blanks, so its denominator differs from the demographic table., begin with participant or material flow, descriptive data and the prespecified primary result before secondary or exploratory findings.

Pair every estimate with enough context to interpret it

Name groups, denominators, units, time points, effect estimates and uncertainty, and distinguish unadjusted from adjusted analysis. Preserve this limit: Do not silently change denominators or impute values without reporting the method.

Use tables and figures to carry evidence without duplication

Give each display a self-contained title, labels, notes and denominators, then use prose for the main pattern rather than reading every cell aloud.

Separate results from explanation and keep deviations visible

Report missing data, exclusions, harms, sensitivity analyses and departures from plan before interpretation. Do not silently change denominators or impute values without reporting the method. Clear reporting supports appraisal; it does not make an estimate causal or important by itself.

Student, parent and career transfer

Practise “Describe Missing Data and Analysis Populations” by separating evidence, interpretation, decision and next action. A parent, teacher, supervisor or teammate can ask what supports the sentence and which missing fact could change it. This discipline transfers to school investigations, university research, laboratory reports, social science, health research, evaluation, data analysis, dissertations, evidence synthesis and research communication without guaranteeing grades, approval, publication, system safety, research validity, employment or future outcomes.

CHAPTER 4 OF 12

4. Present Descriptive Data Before Comparisons

Back to contents

Why this move matters

Describe the analysed sample or materials with units, categories and useful distributions.

Report what was found in the order the study asked

Describe the analysed sample or materials with units, categories and useful distributions. For The report shows programme year, study load and median weekly planning frequency with a suitable spread., begin with participant or material flow, descriptive data and the prespecified primary result before secondary or exploratory findings.

Pair every estimate with enough context to interpret it

Name groups, denominators, units, time points, effect estimates and uncertainty, and distinguish unadjusted from adjusted analysis. Preserve this limit: Baseline description should not become a significance contest.

Use tables and figures to carry evidence without duplication

Give each display a self-contained title, labels, notes and denominators, then use prose for the main pattern rather than reading every cell aloud.

Separate results from explanation and keep deviations visible

Report missing data, exclusions, harms, sensitivity analyses and departures from plan before interpretation. Baseline description should not become a significance contest. Clear reporting supports appraisal; it does not make an estimate causal or important by itself.

Student, parent and career transfer

Practise “Present Descriptive Data Before Comparisons” by separating evidence, interpretation, decision and next action. A parent, teacher, supervisor or teammate can ask what supports the sentence and which missing fact could change it. This discipline transfers to school investigations, university research, laboratory reports, social science, health research, evaluation, data analysis, dissertations, evidence synthesis and research communication without guaranteeing grades, approval, publication, system safety, research validity, employment or future outcomes.

CHAPTER 5 OF 12

5. State the Primary Outcome Clearly

Back to contents

Why this move matters

Report the prespecified primary result with group, time point, denominator, unit and direction.

Report what was found in the order the study asked

Report the prespecified primary result with group, time point, denominator, unit and direction. For The median weekly planning frequency is reported for the full analysed sample before subgroup comparisons., begin with participant or material flow, descriptive data and the prespecified primary result before secondary or exploratory findings.

Pair every estimate with enough context to interpret it

Name groups, denominators, units, time points, effect estimates and uncertainty, and distinguish unadjusted from adjusted analysis. Preserve this limit: A primary label must follow the plan, not the smallest p-value.

Use tables and figures to carry evidence without duplication

Give each display a self-contained title, labels, notes and denominators, then use prose for the main pattern rather than reading every cell aloud.

Separate results from explanation and keep deviations visible

Report missing data, exclusions, harms, sensitivity analyses and departures from plan before interpretation. A primary label must follow the plan, not the smallest p-value. Clear reporting supports appraisal; it does not make an estimate causal or important by itself.

Student, parent and career transfer

Practise “State the Primary Outcome Clearly” by separating evidence, interpretation, decision and next action. A parent, teacher, supervisor or teammate can ask what supports the sentence and which missing fact could change it. This discipline transfers to school investigations, university research, laboratory reports, social science, health research, evaluation, data analysis, dissertations, evidence synthesis and research communication without guaranteeing grades, approval, publication, system safety, research validity, employment or future outcomes.

CHAPTER 6 OF 12

6. Give Effect Estimates and Precision

Back to contents

Why this move matters

Pair differences, ratios or other effects with confidence or credible intervals and relevant units rather than reporting thresholds alone.

Report what was found in the order the study asked

Pair differences, ratios or other effects with confidence or credible intervals and relevant units rather than reporting thresholds alone. For The comparison reports an estimated median difference and its interval under the prespecified method., begin with participant or material flow, descriptive data and the prespecified primary result before secondary or exploratory findings.

Pair every estimate with enough context to interpret it

Name groups, denominators, units, time points, effect estimates and uncertainty, and distinguish unadjusted from adjusted analysis. Preserve this limit: An interval is not a guarantee that the true value lies inside it.

Use tables and figures to carry evidence without duplication

Give each display a self-contained title, labels, notes and denominators, then use prose for the main pattern rather than reading every cell aloud.

Separate results from explanation and keep deviations visible

Report missing data, exclusions, harms, sensitivity analyses and departures from plan before interpretation. An interval is not a guarantee that the true value lies inside it. Clear reporting supports appraisal; it does not make an estimate causal or important by itself.

Student, parent and career transfer

Practise “Give Effect Estimates and Precision” by separating evidence, interpretation, decision and next action. A parent, teacher, supervisor or teammate can ask what supports the sentence and which missing fact could change it. This discipline transfers to school investigations, university research, laboratory reports, social science, health research, evaluation, data analysis, dissertations, evidence synthesis and research communication without guaranteeing grades, approval, publication, system safety, research validity, employment or future outcomes.

CHAPTER 7 OF 12

7. Distinguish Unadjusted and Adjusted Analyses

Back to contents

Why this move matters

Name the model, adjustment variables and rationale previously specified, and present both forms where appropriate.

Report what was found in the order the study asked

Name the model, adjustment variables and rationale previously specified, and present both forms where appropriate. For The section labels the raw comparison and a prespecified adjusted estimate separately., begin with participant or material flow, descriptive data and the prespecified primary result before secondary or exploratory findings.

Pair every estimate with enough context to interpret it

Name groups, denominators, units, time points, effect estimates and uncertainty, and distinguish unadjusted from adjusted analysis. Preserve this limit: Adjustment does not automatically remove confounding or create causality.

Use tables and figures to carry evidence without duplication

Give each display a self-contained title, labels, notes and denominators, then use prose for the main pattern rather than reading every cell aloud.

Separate results from explanation and keep deviations visible

Report missing data, exclusions, harms, sensitivity analyses and departures from plan before interpretation. Adjustment does not automatically remove confounding or create causality. Clear reporting supports appraisal; it does not make an estimate causal or important by itself.

Student, parent and career transfer

Practise “Distinguish Unadjusted and Adjusted Analyses” by separating evidence, interpretation, decision and next action. A parent, teacher, supervisor or teammate can ask what supports the sentence and which missing fact could change it. This discipline transfers to school investigations, university research, laboratory reports, social science, health research, evaluation, data analysis, dissertations, evidence synthesis and research communication without guaranteeing grades, approval, publication, system safety, research validity, employment or future outcomes.

CHAPTER 8 OF 12

8. Report Categories, Cut Points and Continuous Measures

Back to contents

Why this move matters

Explain how continuous variables were grouped and report continuous results when the plan requires them.

Report what was found in the order the study asked

Explain how continuous variables were grouped and report continuous results when the plan requires them. For Study load categories use published hour ranges and the original continuous hours remain available in a table., begin with participant or material flow, descriptive data and the prespecified primary result before secondary or exploratory findings.

Pair every estimate with enough context to interpret it

Name groups, denominators, units, time points, effect estimates and uncertainty, and distinguish unadjusted from adjusted analysis. Preserve this limit: Data-driven cut points can exaggerate apparent patterns.

Use tables and figures to carry evidence without duplication

Give each display a self-contained title, labels, notes and denominators, then use prose for the main pattern rather than reading every cell aloud.

Separate results from explanation and keep deviations visible

Report missing data, exclusions, harms, sensitivity analyses and departures from plan before interpretation. Data-driven cut points can exaggerate apparent patterns. Clear reporting supports appraisal; it does not make an estimate causal or important by itself.

Student, parent and career transfer

Practise “Report Categories, Cut Points and Continuous Measures” by separating evidence, interpretation, decision and next action. A parent, teacher, supervisor or teammate can ask what supports the sentence and which missing fact could change it. This discipline transfers to school investigations, university research, laboratory reports, social science, health research, evaluation, data analysis, dissertations, evidence synthesis and research communication without guaranteeing grades, approval, publication, system safety, research validity, employment or future outcomes.

CHAPTER 9 OF 12

9. Design Tables and Figures as Evidence

Back to contents

Why this move matters

Use self-contained titles, units, denominators, legends and notes, and make the display accessible and readable.

Report what was found in the order the study asked

Use self-contained titles, units, denominators, legends and notes, and make the display accessible and readable. For A figure labels every axis and the table footnote explains missing denominators and adjusted estimates., begin with participant or material flow, descriptive data and the prespecified primary result before secondary or exploratory findings.

Pair every estimate with enough context to interpret it

Name groups, denominators, units, time points, effect estimates and uncertainty, and distinguish unadjusted from adjusted analysis. Preserve this limit: Colour alone should not carry group meaning.

Use tables and figures to carry evidence without duplication

Give each display a self-contained title, labels, notes and denominators, then use prose for the main pattern rather than reading every cell aloud.

Separate results from explanation and keep deviations visible

Report missing data, exclusions, harms, sensitivity analyses and departures from plan before interpretation. Colour alone should not carry group meaning. Clear reporting supports appraisal; it does not make an estimate causal or important by itself.

Student, parent and career transfer

Practise “Design Tables and Figures as Evidence” by separating evidence, interpretation, decision and next action. A parent, teacher, supervisor or teammate can ask what supports the sentence and which missing fact could change it. This discipline transfers to school investigations, university research, laboratory reports, social science, health research, evaluation, data analysis, dissertations, evidence synthesis and research communication without guaranteeing grades, approval, publication, system safety, research validity, employment or future outcomes.

CHAPTER 10 OF 12

10. Use Prose for the Pattern, Not Every Cell

Back to contents

Why this move matters

Point readers to the central comparison, direction and uncertainty without duplicating the whole table.

Report what was found in the order the study asked

Point readers to the central comparison, direction and uncertainty without duplicating the whole table. For The paragraph names the primary difference and refers to Table 2 for full subgroup estimates., begin with participant or material flow, descriptive data and the prespecified primary result before secondary or exploratory findings.

Pair every estimate with enough context to interpret it

Name groups, denominators, units, time points, effect estimates and uncertainty, and distinguish unadjusted from adjusted analysis. Preserve this limit: Selective narration must not hide a contradictory result.

Use tables and figures to carry evidence without duplication

Give each display a self-contained title, labels, notes and denominators, then use prose for the main pattern rather than reading every cell aloud.

Separate results from explanation and keep deviations visible

Report missing data, exclusions, harms, sensitivity analyses and departures from plan before interpretation. Selective narration must not hide a contradictory result. Clear reporting supports appraisal; it does not make an estimate causal or important by itself.

Student, parent and career transfer

Practise “Use Prose for the Pattern, Not Every Cell” by separating evidence, interpretation, decision and next action. A parent, teacher, supervisor or teammate can ask what supports the sentence and which missing fact could change it. This discipline transfers to school investigations, university research, laboratory reports, social science, health research, evaluation, data analysis, dissertations, evidence synthesis and research communication without guaranteeing grades, approval, publication, system safety, research validity, employment or future outcomes.

CHAPTER 11 OF 12

11. Label Secondary, Sensitivity and Exploratory Analyses

Back to contents

Why this move matters

Identify analysis status, rationale and multiplicity or robustness context so discovery is not presented as confirmation.

Report what was found in the order the study asked

Identify analysis status, rationale and multiplicity or robustness context so discovery is not presented as confirmation. For A post-hoc comparison by commuting time is explicitly exploratory., begin with participant or material flow, descriptive data and the prespecified primary result before secondary or exploratory findings.

Pair every estimate with enough context to interpret it

Name groups, denominators, units, time points, effect estimates and uncertainty, and distinguish unadjusted from adjusted analysis. Preserve this limit: An interesting exploratory pattern still needs future confirmation.

Use tables and figures to carry evidence without duplication

Give each display a self-contained title, labels, notes and denominators, then use prose for the main pattern rather than reading every cell aloud.

Separate results from explanation and keep deviations visible

Report missing data, exclusions, harms, sensitivity analyses and departures from plan before interpretation. An interesting exploratory pattern still needs future confirmation. Clear reporting supports appraisal; it does not make an estimate causal or important by itself.

Student, parent and career transfer

Practise “Label Secondary, Sensitivity and Exploratory Analyses” by separating evidence, interpretation, decision and next action. A parent, teacher, supervisor or teammate can ask what supports the sentence and which missing fact could change it. This discipline transfers to school investigations, university research, laboratory reports, social science, health research, evaluation, data analysis, dissertations, evidence synthesis and research communication without guaranteeing grades, approval, publication, system safety, research validity, employment or future outcomes.

CHAPTER 12 OF 12

12. Report Deviations, Harms and Null Findings Honestly

Back to contents

Why this move matters

Describe departures from plan, unintended events where relevant, analyses that could not be completed and results that do not support the hypothesis.

Report what was found in the order the study asked

Describe departures from plan, unintended events where relevant, analyses that could not be completed and results that do not support the hypothesis. For The report states that one planned subgroup was too small for the specified analysis and does not replace it quietly., begin with participant or material flow, descriptive data and the prespecified primary result before secondary or exploratory findings.

Pair every estimate with enough context to interpret it

Name groups, denominators, units, time points, effect estimates and uncertainty, and distinguish unadjusted from adjusted analysis. Preserve this limit: Absence of statistical significance is not proof of no effect.

Use tables and figures to carry evidence without duplication

Give each display a self-contained title, labels, notes and denominators, then use prose for the main pattern rather than reading every cell aloud.

Separate results from explanation and keep deviations visible

Report missing data, exclusions, harms, sensitivity analyses and departures from plan before interpretation. Absence of statistical significance is not proof of no effect. Clear reporting supports appraisal; it does not make an estimate causal or important by itself.

Student, parent and career transfer

Practise “Report Deviations, Harms and Null Findings Honestly” by separating evidence, interpretation, decision and next action. A parent, teacher, supervisor or teammate can ask what supports the sentence and which missing fact could change it. This discipline transfers to school investigations, university research, laboratory reports, social science, health research, evaluation, data analysis, dissertations, evidence synthesis and research communication without guaranteeing grades, approval, publication, system safety, research validity, employment or future outcomes.

PRACTICE LAB

Build a research results section decision trail

Student route

1. Choose one real example or official document inside this guide’s scope.
2. Copy its objective and underline actor, action, context, evidence and intended decision.
3. Draw the specific reasoning chain used in this article.
4. Mark each step reported, inferred, contested or unknown.
5. Rewrite the conclusion so it preserves the narrowest important limitation.

Parent or teacher route

Ask the learner to explain turn analysed data into a transparent results section that shows who and what was studied, what the prespecified analyses found, how uncertain the estimates are and where readers can verify the evidence twice: once to a classmate and once to a method specialist. Praise a precise correction and a well-stated unknown; both show disciplined attention to evidence.

Early-career route

Create a one-page research results section audit with source link, date checked, scope, method logic, decisive wording, contrary evidence, limit and next action. A colleague should be able to continue without guessing.

Continue with eduKate’s guides to reading course and career information and giving useful feedback in group projects.

USEFUL NEXT READING

Connect this narrow owner to its neighbour

Read the research methods-section guide for a related but different reader job. Return here when the question is specifically how to turn analysed data into a transparent results section that shows who and what was studied, what the prespecified analyses found, how uncertain the estimates are and where readers can verify the evidence. Use each guide for the reader task named in its heading.

For English practice across research, school and career communication, visit the Singapore English Tuition Centre learning site.

FREQUENTLY ASKED QUESTIONS

Questions students and parents often ask

Should the results section include interpretation?

Report the finding and necessary factual context; broader explanation, implications and comparison with literature usually belong in the discussion.

Must I report p-values?

Follow the study plan and field guidance, but do not let a threshold replace effect size, uncertainty and context.

Can tables repeat the text?

Use the text for the main pattern and tables for complete detail; avoid copying every number into prose.

What if denominators differ?

State the denominator for each result and explain missingness or exclusions so percentages remain interpretable.

How do I report a null result?

Report the estimate and uncertainty accurately without claiming proof of no effect merely because a threshold was not crossed.

Does following STROBE or CONSORT prove the study is valid?

No. Reporting guidelines improve completeness; design, conduct, analysis and inference still require appraisal.