Methods sections often name an approach without showing how decisions shaped the result. This module asks learners to show the route from data to finding. Quantitative studies need clear models, tests, assumptions, uncertainty, and analytic decisions. Qualitative studies need context, researcher position, coding, interpretation, and an account of how themes or claims were developed. Mixed-method studies need both strands and their connection.
Map a quantitative claim from the analysed data through transformations, tests or models, and uncertainty estimates to its interpretation.
Document qualitative context, researcher position, coding, analytic development, and the evidence supporting an interpretive claim.
Distinguish primary, secondary, and exploratory analyses when describing quantitative findings.
Explain how qualitative and quantitative strands were connected in a mixed-methods study.
A paper reports that an intervention improved learning, names a regression model, and gives a p-value, but does not state which cases were excluded, how missing data were handled, whether assumptions were examined, or how the model changed from the initial plan. A reader cannot check how the reported result was produced. The same problem occurs when a qualitative paper names “thematic analysis” but does not explain the study context, researchers’ roles, coding process, or how extracts support an interpretation.
Transparent analysis reporting makes the route from evidence to conclusion inspectable. It helps readers assess what was done, what decisions shaped the finding, and where uncertainty or limits remain. Reporting detail does not by itself make an analysis correct or guarantee that another researcher will reproduce a finding; it makes the basis for evaluation clearer.
For every reported claim, give readers a traceable route:
Data: What material was analysed, and which participants, records, documents, observations, or measurements contributed?
Preparation: What processing occurred before analysis—for example, exclusions, treatment of missing information, transformations, transcription, anonymisation, or coding?
Analysis: What test, model, coding procedure, comparison, or interpretive process was used?
Finding: What result, theme, pattern, or inference emerged?
Interpretation: What does the finding mean in relation to the research question, context, and limits of the study?
This route is not a demand that all studies use the same format. Quantitative, qualitative, and mixed-methods studies use different forms of evidence and reasoning, but each should make its analytic decisions visible enough for readers to understand and assess the claims. Qualitative reporting can use narrative, thematic, chronological, or other forms that fit the inquiry, provided the necessary information is present.
When writing or reviewing, work backwards from each major claim. Ask: What data support this? What happened to the data before analysis? What analytic procedure connected the data to this claim? What interpretation goes beyond the immediate result?
Name the test or model used and explain its role. A test evaluates a specified statistical question; a model represents relationships among variables using stated assumptions. For complex analyses, readers need enough detail to know what model was estimated, what variables were included, and what analytic problems or changes occurred.
Report the data decisions that shaped the analysis. State planned criteria for excluding cases, how missing data were addressed, how outliers were defined and processed, whether distributions were examined, and whether transformations were applied. If decisions changed after analysis began, identify the change and give its rationale rather than presenting it as though it had always been planned.
Report whether important assumptions—conditions required for an analysis to be appropriate—were examined, and describe problems that could affect the validity of findings. Do not imply that an assumption check is a ritual pass/fail step; explain what was found and how it affected the analysis or interpretation.
Pair statistical tests with estimates of magnitude and uncertainty. An effect size describes the size of a difference, association, or other estimated relationship. A confidence interval gives an interval estimate around an effect size; report it alongside the estimate when possible.
If you use null-hypothesis statistical testing, report the relevant p-values and the information needed to understand the test. A p-value alone does not show practical importance or the probability that a hypothesis is true. Separate the statistical result from the substantive interpretation, including why the observed magnitude matters—or may not matter—in the study context.
Label analyses honestly. Identify which hypotheses and analyses were primary (central to the study), secondary (additional planned questions), and exploratory (analyses used to investigate patterns beyond the planned focus). Exploratory analysis can be valuable, but readers need to know its status when judging how strongly a result supports a conclusion.
Begin with context. Explain the setting, relevant features of participants or data sources, and the conditions in which data were generated or selected. Context is not background decoration: it helps readers understand what the data represent and the boundaries within which an interpretation is meaningful.
Describe the researchers’ position in relation to the study where it could shape data collection or interpretation. This may include relevant experience, relationships with participants, assumptions, or commitments, as well as how these were managed or used in the inquiry. Reflexivity is not a scripted personal disclosure; it is an account of how researcher perspectives mattered to the analytic process.
Make coding and analysis visible. Define coding as assigning labels to portions of data to support systematic interpretation. Explain whether codes were developed before analysis, emerged during analysis, or changed iteratively; identify the unit of analysis; describe who analysed the data; and show how themes, categories, narratives, or other inferences developed.
Do not reduce qualitative quality to a list of statistical-style checks. Different qualitative approaches have distinct assumptions, procedures, and forms of presentation, and reports should explain choices in relation to the study’s goals and approach. Transparent reporting helps readers evaluate the work, but a reporting framework should not be used as a rigid judgement of methodological quality.
Link interpretations to empirical material. Use appropriately contextualised quotations, field notes, text excerpts, observations, or other evidence to substantiate main analytic findings, while also explaining how the extract contributes to the claim. Extracts illustrate and ground an analysis; they do not replace the author’s account of the interpretation.
A mixed-methods report should make both strands visible: the quantitative data, preparation, models or tests, and results; and the qualitative context, analytic procedures, interpretations, and supporting extracts. It should also explain why both forms of evidence were needed for the research aim.
Then show how the strands were joined. State whether results were merged, connected in sequence, compared, or used to build the next phase of the study. Present the integrated finding—not merely two parallel result sections—and explain whether the strands converged, extended one another, or revealed a meaningful tension. Tables or other joint displays can place qualitative findings beside quantitative results so readers can see the basis for the integration.
Claim: Students who received the learning prompt showed higher post-course engagement scores than students in the comparison condition.
Evidence trail a reader should be able to follow:
Data: The final analysed sample in each condition, participant flow, and the engagement measure.
Preparation: Any exclusions, missing-data decisions, outlier handling, and transformations.
Analysis: The named comparison test or model, included variables, and relevant assumptions or diagnostic findings.
Finding: The estimated group difference, its effect size, confidence interval, and p-value where null-hypothesis testing was used.
Interpretation: A statement distinguishing the estimated difference from a claim about practical educational importance, with relevant limitations.
Quantitative reports should provide results for inferential tests, effect-size estimates, and confidence intervals when possible.
Claim: Participants described the prompt as useful when it helped them notice gaps in their study planning.
Evidence trail a reader should be able to follow:
Data and context: Who contributed the interviews or other material, the learning setting, and relevant contextual features.
Researcher position: Any researcher-participant relationship, prior understandings, or reflexive practices relevant to interpretation.
Preparation and coding: How recordings or texts were prepared, what was coded, how codes developed, and how the theme was formed.
Finding: The theme, its interpretation, and extracts showing participants’ accounts of noticing planning gaps.
Interpretation: The scope of the claim, including what the evidence suggests in this context and what it does not establish beyond it.
Qualitative reporting should show both the analytic finding and the empirical material that substantiates it.