Reporting research and course review
Introduction to Social Research Methodology
Reporting research to its audiences
The last link in the chain
Reporting is the sixth stage of the research process mapped in session 1, and it is the stage at which studies most often fail the people they were meant to serve. Research that is never communicated, or communicated badly, might as well not have been done: the quality of the design, the sampling, and the analysis counts for nothing if the findings never reach a decision-maker in a form they can use. For the Meridian research team, this is the moment when months of evidence about turnover and customer satisfaction must become something the board can act on.
One study normally produces several documents, because it has several audiences. A full research report provides the complete, permanent record; a managerial presentation supports the decision; an executive summary serves the many readers who will only ever read one page. Good reporting is therefore a design problem: you start from the audience and work backwards, rather than starting from the study and hoping the audience keeps up.
The structure of a research report
The report is the study’s archive. It should be complete enough that a sceptical reader can check every claim against the evidence behind it. A typical organisational research report contains, in order:
| Section | What it contains |
|---|---|
| Title page and executive summary | The one-page distillation, findings first (see below) |
| Background and aims | The problem, its context, and the research question |
| Methods | Design, sampling, and data collection, in enough detail to be scrutinised |
| Findings | The results, organised by question or theme — not by the order things happened |
| Discussion | Interpretation of the findings, and the study’s limitations |
| Conclusions and recommendations | Clearly separated and clearly labelled |
| Appendices | Instruments, full tables, and technical detail |
The academic version of the same logic is the IMRaD structure — Introduction, Methods, Results, and Discussion — which organises most journal articles. The order and emphasis differ slightly, but the underlying commitment is the same: the reader must be able to see not only what was found, but how.
The structure of a managerial presentation
A presentation is not the report projected onto slides; it has its own architecture, built around the fact that board attention is scarce. Where the report proceeds from question through methods to findings, the presentation inverts the order: the answer comes first, and the evidence follows. This is often called the “pyramid principle”. A sound managerial presentation runs:
- The question the board asked.
- The answer — the headline findings.
- The key evidence behind each finding.
- The recommendations, with options and costs.
- Methods in reserve — backup slides prepared and ready for questions.
Methodology gets one slide at most in the main flow. The detail exists — the presenter must be able to defend every claim — but it waits to be asked for rather than consuming the audience’s attention up front.
Report vs presentation: a comparison
| Research report | Managerial presentation | |
|---|---|---|
| Purpose | Record and scrutiny | Decision |
| Order | Question → methods → findings | Answer first, evidence after |
| Detail | Complete | Ruthlessly selective |
| Methods | A full chapter | One slide, plus reserve slides |
| Length | Read at the reader’s own pace | 10–15 minutes of scarce attention |
Both documents rest on the same analysis. What changes between them is the order and the selection of material — never the honesty of the content. A finding that is inconvenient does not become optional because the audience is senior.
The executive summary
The executive summary is a craft of its own — the hardest writing per word in the whole report. Its rules are strict:
- Findings first. It opens with what was found, not with the history of the project or a lecture on methodology.
- One page at most, and self-contained: many readers, including some of the most important ones, will read nothing else.
- Plain language, the few numbers that genuinely matter, and recommendations clearly labelled as recommendations rather than smuggled in among the findings.
- Written last, even though it appears first — you cannot summarise a study that is not yet finished.
The test of a good executive summary: could a board member who reads nothing else make a defensible decision from this page alone?
Tailoring to the audience
The same findings typically reach at least three audiences, and each needs a different document:
| Audience | Orientation | What they need |
|---|---|---|
| The board | Decision-focused | The answer, the implications, the costs and risks; minimal method |
| The academic reader | Method-focused | Design, measures, and analysis, in enough detail to judge whether the claims are warranted |
| All-staff communication | Trust-focused | What we found, what will change, and what happened to your data |
What varies across these documents is order, detail, and vocabulary. What must never vary is the findings themselves: tailoring is a matter of presentation, not of telling each audience what it wants to hear. Reporting back to the employees who answered the survey also closes the loop with respondents — and protects the response rate of every future survey the organisation runs. At Meridian, the staff who completed the engagement survey should hear the results before they read about the changes in a press release.
Principles of accurate reporting
Report uncertainty honestly
Every estimate from a sample carries uncertainty, and accurate reporting shows that uncertainty rather than hiding it. For survey estimates, report the margin of error or confidence interval — “64%, plus or minus 5 points” — so that small differences between groups are not mistaken for real ones. Report the response rate, and say who is likely to be missing: non-response bias matters, and the people you most need to hear from (at Meridian, employees who have already left) are usually the hardest to reach. In qualitative work the same honesty takes a different form: report how many people were studied, how they were selected, and what settings the findings can plausibly speak to.
Avoid false precision. “About two-thirds” is a more honest description of a result from a sample of 300 than “66.42%”, whose decimal places imply an accuracy the data cannot support. A figure such as “engagement is 3.6 in logistics and 3.9 in stores” means little until the reader also knows the sample sizes and the uncertainty around each number.
Limitations are part of the findings
Session 1 closed with the point that acknowledging limitations is a mark of good research, not a confession of failure; the reporting stage is where that principle reaches the page. Limitations belong in the discussion section — visible, not buried in an appendix — and should be echoed wherever a specific claim depends on them. The usual suspects are:
- Coverage — who was missed by the sampling or the mode of data collection.
- Measurement — self-report, social desirability, and imperfect operationalisations.
- Timing — a single point in time cannot show change or direction.
- Generalisability — which settings, populations, and periods the results can travel to.
A report that claims no limitations is not a stronger report; it is a less honest one. Stated limitations also do productive work: they define what the next study should examine, which is why good reports pair them with recommendations for further research.
Findings are not recommendations
A finding is what the evidence showed. A recommendation is what the organisation should do about it. The recommendation always adds something the data cannot supply: judgements about values, costs, priorities, and risk. The classic reporting failure is to slide from one to the other without noticing; accurate reports label the two and keep them in separate sections.
The Meridian case illustrates the gap. Finding: turnover is concentrated among first-year store staff, who cite unpredictable scheduling. Recommendation: pilot fixed schedules in five stores. The second does not follow automatically from the first — it prices in implementation costs, operational feasibility, and the board’s appetite for risk. Evidence narrows the options; it rarely dictates the choice.
Do not overclaim causality
A cross-sectional survey establishes association, not causation — however managerially convenient a causal story would be. The damage is usually done by language: verbs such as “drives”, “leads to”, and “because of” attached to merely correlational results. Honest wording uses “is associated with”, “predicts”, or “is consistent with”, and explicitly considers what else could produce the pattern: reverse causation and confounding variables. A defensible causal claim wants an experimental or at least a longitudinal design.
Meridian again: stores with low engagement have high turnover. Does low engagement cause quitting? Does constant churn depress the engagement of those who remain? Or does a weak store manager cause both? The cross-sectional survey cannot say — and the report must not pretend that it can.
Report what did not support expectations
Reporting only the results you hoped for — cherry-picking — distorts the overall picture as surely as inventing data would. Null and negative results are findings: at Meridian, “pay was not the main driver of turnover” redirects real money away from an across-the-board raise that would not have solved the problem. Accurate practice is to report every analysis that was planned, and to label exploratory, after-the-fact analyses as exploratory rather than presenting them as if they had been predicted.
The pressure to tell the sponsor what it wants to hear is real in organisational research, where the sponsor also pays the invoice. Resisting that pressure is precisely what makes the research worth paying for. The board expected pay to explain the turnover spike; the data pointed to scheduling and supervision instead. That surprise is the single most valuable finding in the study.
Visualising results honestly
One message per chart
A chart is an argument, not decoration: know the message before you design the figure. Put the message in the title — “Turnover is concentrated in first-year store staff”, not “Figure 3: Turnover data”. Hold to one comparison per chart: if a figure needs a paragraph of explanation, split it into two figures or simplify it. The governing question for every chart is: what comparison is the reader supposed to make, and does the design make that comparison easy?
Choose the chart for the comparison
The comparison dictates the form:
| You want to show | Reach for | Avoid |
|---|---|---|
| Comparison across categories | Bar chart, axis from zero | Truncated bars, 3-D effects |
| Trend over time | Line chart | Extrapolating beyond the data |
| Parts of a whole | Pie chart (two or three shares only) or stacked bar | Many-slice pies |
| Relationship between two variables | Scatterplot | Implying causation |
| Exact values | A small, well-formatted table | Charts that make readers guess |
Two of these deserve emphasis. Pie charts work only for two or three shares, because humans compare bar lengths far more accurately than angles; beyond that, use bars. And when the exact values matter more than the pattern, a table beats a chart.
Honest axes
Bar charts must start at zero. A bar encodes its value by its length, so a truncated axis makes the bar lie: engagement scores of 3.9 and 3.6 plotted on an axis starting at 3.0 turn an 8% difference in the numbers into bars that differ in height by 50%. Line charts may legitimately zoom in on a range to show change — but the axis must be clearly labelled and the zoom acknowledged. Never extrapolate a line beyond the data as if the projection were observed, and keep scales consistent across charts the reader is meant to compare. The test of an honest chart is simple: does the visual impression match what the numbers actually say?
Declutter and label directly
Every element of a chart must earn its place. Drop heavy gridlines, borders, backgrounds, and decorative colour; never use 3-D effects, which distort every length and angle they touch. Label directly — put values on or beside the bars, and series names at the ends of lines — instead of forcing the reader to shuttle back and forth to a legend. Use colour to carry meaning: highlight the one bar the message is about and mute the rest, rather than giving every category its own festive hue. Round the numbers used as labels, and put the source, the sample size, and the scale in a footnote. The four-point check for every chart: message in the title, honest axes, no clutter, labels where the eyes already are.
Course review and the examination
One connected research process
The course has been a single journey in disguise — the research process introduced in session 1, taken one stage at a time:
| Stage | Sessions |
|---|---|
| From problem to research question | 2 |
| Hypotheses and research design | 3 |
| Research ethics | 4 |
| Literature review and secondary data | 5–6 |
| Choosing approaches: quantitative, qualitative, mixed | 7–9 |
| Sampling and recruitment | 10 |
| Collecting data | 11–12 |
| Analysing data | 13–14 |
| Reporting research | 15 (today) |
The chain metaphor from session 1 holds to the end: each choice constrains the next, and a study is only as strong as its weakest link. Meridian has been the thread running through the course; by now you can take a board’s vague worry — “people keep leaving and we don’t know why” — and turn it into a designed, sampled, collected, analysed, and reported diagnosis.
The distinctions that carry the course
If you hold the following distinctions securely — and, crucially, can apply them to a scenario — you are ready for the examination:
| Distinction | The core of it |
|---|---|
| Qualitative vs quantitative | Meaning and depth vs measurement and breadth — and matching the method to the question |
| Induction vs deduction | Building theory up from observations vs testing hypotheses derived from theory |
| Probability vs non-probability sampling | Known chances of selection licensing generalisation vs convenience and criteria that do not |
| Validity vs reliability | Accuracy (measuring what you intend) vs consistency (getting the same result again) |
| Correlation vs causation | Association vs a defensible causal claim, which wants experimental or longitudinal evidence |
| Findings vs recommendations | What the evidence showed vs what the organisation should do about it |
Every one of these has appeared at least twice in the course, because every one of them recurs constantly in real research.
The examination: format and grading
The examination is written, and consists of open and/or closed questions. It is scored out of 100 points, and the pass mark is 51. Closed questions test recognition of core concepts; open questions test whether you can explain and apply them. Everything examinable was covered in the sessions and is summarised in the handouts.
| Points | Grade |
|---|---|
| 91–100 | 5 (very good) |
| 81–90 | 4+ (good plus) |
| 71–80 | 4 (good) |
| 61–70 | 3+ (satisfactory plus) |
| 51–60 | 3 (satisfactory) |
| Below 51 | 2 (fail — the course must be repeated) |
How to revise
- The handouts are your study guides. Each one was written to summarise its session completely, with every definition, typology, and distinction you need. Revising the handouts covers the examinable material.
- Understanding beats memorising. Expect application questions of the form “which design fits this scenario, and why?” or “which sampling strategy is this, and what can it support?”. A memorised definition that cannot be applied earns little.
- Reconstruct, don’t reread. For each comparison table in the handouts, close the document and rebuild the table from memory, then check what you missed. Active reconstruction is far more effective than passive rereading.
- Attach your own examples. For every concept, invent an organisational example of your own — an example you built is far more retrievable under exam conditions than one you read.
- Practise on scenarios. The revision-circuit exercise in this session is a rehearsal for exactly the kind of thinking the examination rewards.
Example exam-style questions
- Closed. What is a key characteristic of non-probability sampling? (a) it guarantees a representative sample; (b) participants are selected by convenience or accessibility; (c) it requires a complete list of the population; (d) it is used mainly for large-scale surveys. — The answer is (b).
- Closed. What is the primary aim of the explanatory sequential mixed-methods design? (a) to generate quantitative data from qualitative findings; (b) to run both strands simultaneously; (c) to use qualitative data to explain quantitative results; (d) to replicate quantitative findings. — The answer is (c).
- Open. A company’s survey shows that stores with lower engagement have higher turnover. Explain why this alone cannot establish that low engagement causes turnover, and name one design that could support a causal claim. — A full answer names the cross-sectional nature of the evidence, reverse causation, and confounding, and proposes a longitudinal (panel) or experimental design.
- Open. A firm’s customer satisfaction is falling. Plan a small study to diagnose why: state a research question, a design, a sampling strategy, and one limitation. — A full answer makes each choice and gives a reason for it; the reasons carry most of the marks.
Conclusion
Reporting completes the research process, and it is governed by the same honesty as every earlier stage. The report is the complete record; the presentation puts the answer first and holds the methods in reserve; the executive summary is a one-page craft that leads with findings. Accurate reporting shows uncertainty, owns its limitations, separates findings from recommendations, refuses to overclaim causality, and reports the results nobody hoped for. Honest visualisation chooses the chart to fit the comparison, starts bars at zero, carries one message per figure, and labels it directly.
Looking back, the course was one connected chain — problem, question, design, ethics, literature, approach, sample, data, analysis, report — and the examination rewards those who can apply its distinctions to new scenarios, not merely recite them. Revise from the handouts, reconstruct the tables, build your own examples, and practise on scenarios. The pass mark is 51 out of 100; aim comfortably above it.