Before and after observation, two cafes

Undergraduate research poster

An undergraduate research poster does not need big numbers, it needs honest ones arranged in the standard scientific order. This A0 portrait example shows a four week audit of a 50 cent reusable cup discount at two campus cafes: 4,860 sales observed, use rising from 12.4 to 29.1 percent, and a conclusion that explicitly declines to claim causation.

Create a poster with OneCraftA0 Portrait, printed at 841 by 1189 millimetres

The whole board

The poster at full size, exactly as it prints. Every number, citation and caption on it was written for this example, so the layout is being judged on real content.

KU
A 50 cent discount raised reusable cup use from 12 to 29 percent at two campus cafes
L. Farrow ¹, P. Quist (supervisor) ¹
1 School of Environmental Science, Kingsbridge University
Abstract
Campus sustainability plans lean heavily on reusable cups and rarely measure whether anything changes behaviour at the till. This audit observed 4,860 hot drink sales at two campus cafes over four weeks, two weeks before and two weeks after a 50 cent discount for any reusable cup was introduced with a poster at the counter. Reusable cups were used for 12.4 percent of sales before the discount (298 of 2,410) and 29.1 percent after (713 of 2,450), a rise of 16.7 percentage points (95% CI 14.5 to 18.9, p < 0.001). Both cafes moved the same way and by a similar amount. Because there was no control cafe, the design cannot separate the discount from the poster or the weather, and this board claims an association, not a cause. The result is still useful: it prices what a term long controlled trial should test.
Background
The university estimates its cafes hand out around 480,000 single use cups a year, and the lids and lining make them effectively unrecyclable in the campus waste stream, so reduction is the only lever that matters. Discount schemes are the standard response, yet published results from cafes elsewhere range from almost no effect to a doubling of reusable use, which is too wide a range to plan on. Most of those studies also relied on till records that only capture a discount actually claimed, missing customers who bring a cup and never ask. This audit was designed to be small, direct and checkable within one term: a person standing at each counter recording what every hot drink was served in, for two weeks either side of the discount starting. The question was not whether a discount can work somewhere, but what it did at these two cafes, counted at the till, with numbers a reader can recompute from this board.
METHODS
Setting. Library Cafe and Science Cafe, weekdays 8 to 11 am, the peak hot drink window at both sites.
Observation. One observer per cafe recorded cup type at the till for every hot drink sold; 2,410 sales before, 2,450 after.
Intervention. A 50 cent discount for any reusable cup from week 3, announced by one poster at each till and nothing else.
Analysis. Chi square on cup type by period; percentage point difference with a 95% confidence interval.
Ethics. Counts only, no customer details recorded; approved as a low risk audit by the school, reference ES/2026/031.
Reusable share (%)
Week 1
Week 2
Week 3
Week 4
08162432
Figure 1. Reusable cup share of hot drink sales by week. The discount began in week 3; before mean 12.4 percent, after mean 29.1.
Source: Trial dataset, 2026.
Results
Reusable cups rose from 12.4 percent of hot drink sales before the discount to 29.1 percent after, a difference of 16.7 percentage points (95% CI 14.5 to 18.9, p < 0.001). The counts behind those percentages are printed so they can be checked: 298 of 2,410 sales before, 713 of 2,450 after. The week by week pattern says more than the averages. Weeks 1 and 2 sat at 11.9 and 12.9 percent, close enough to call a stable baseline. Week 3, the first with the discount, jumped to 26.8 percent, and week 4 rose again to 31.4 percent rather than falling back, so the change did not decay within the audit window; if anything it was still growing as word spread. The two cafes told the same story at different levels. Library Cafe went from 13.1 to 31.6 percent and Science Cafe from 11.7 to 26.5, so the effect was not carried by one site, one barista or one queue culture. Total sales barely moved between periods, 2,410 against 2,450, which matters because a discount could in principle attract new customers rather than change existing ones; a 1.7 percent change in volume cannot explain a 16.7 point change in share. Taken together the numbers describe roughly 205 fewer single use cups per week across the two cafes during the audit, at a cost of about 168 per week in claimed discounts, or roughly 0.82 per cup avoided. That unit cost is the audit's most reusable number, because it is what a control trial has to beat. One detail is worth its sentence: claimed discounts ran slightly below observed reusable cups in both weeks, 671 claims against 713 observed uses in the final fortnight, meaning about six percent of reusable customers never asked for their 50 cents. Counting at the till rather than from till receipts is what caught them, and it is why the percentages on this board are counts of behaviour, not counts of claims.
After (%)
Before (%)
08162432
Library Cafe
Science Cafe
Figure 2. Reusable share before and after the discount, by cafe. Both cafes moved the same way: Library 13.1 to 31.6, Science 11.7 to 26.5.
Source: Trial dataset, 2026.
Discussion
The honest sentence first: nothing in this design proves the discount caused the change. The poster arrived with the discount, the weather warmed across the four weeks, and exam season ended in week 4, so at least three explanations moved together and no control cafe was watched to separate them. The size and shape of the change still constrain the alternatives. A weather or season effect would be expected to drift gradually; what the week chart shows is a step, 12.9 to 26.8 percent, in exactly the week the discount began, at both cafes at once. That is the pattern an intervention leaves, even if this audit cannot prove which part of the intervention did the work, the 50 cents or the poster announcing it. Observation itself is the other limit. The observers were visible, and customers who noticed them may have behaved differently, though it is hard to see why that would change between periods when the observers did not. Counting cup type at the till also says nothing about substitution: a customer who switched from a bought coffee to the free machine in their department disappears from the denominator entirely. Against published cafe studies, a 16.7 point rise sits at the high end, which is plausible for a campus where most customers are repeat daily buyers who can change habit once and keep it. The audit was costed as well as counted: at roughly 0.82 of discount per single use cup avoided, the scheme is cheap against the university procurement cost of the cups themselves, before any waste handling saving. The next step is written into the conclusion rather than implied: a term long trial with a control cafe, till linked counting, and the poster and discount introduced separately.
CONCLUSION
Reusable cup use went from 12.4 to 29.1 percent when a 50 cent discount arrived, both cafes moved together, and the step landed in the exact week the scheme began. Causation is not claimed: a term long trial with a control cafe, and the poster separated from the discount, is the recommendation this board makes.
Reusable share rose from 12.4 to 29.1 percent
Both cafes moved the same way in the same week
No control cafe, so causation is not claimed
Recommend a term long trial with a control cafe
REFERENCES
1.Farrow L, Quist P. Observed cup choice at campus cafes: an audit protocol. Campus Sustainability Notes 2026;4(1):18-27.
2.Quist P, Lindahl F. Price incentives and single use packaging: a short review. Environmental Behaviour Briefs 2024;7(2):66-80.
3.Halloran J, Farrow L. Before and after designs in small field audits: what they can and cannot claim. Student Research Methods 2023;2(3):101-112.

Block by block

What each block on the board is for, in the order a reader walks it.

Title band, crest and authors
The title states the finding in plain numbers, 12 to 29 percent, without a causal verb. The author block carries the student as presenting author and the supervisor with the same affiliation, which is the honest shape of most undergraduate projects.
Abstract and introduction
The abstract prints the counts behind both percentages, 298 of 2,410 and 713 of 2,450, and says outright that the design claims an association, not a cause. The introduction sizes the problem, around 480,000 single use cups a year on campus, and explains why counting at the till beats till receipts.
Methods
Five labelled entries: the two cafes and their morning window, the observation method with both period counts, the intervention, one poster and 50 cents off from week 3, the chi square analysis, and the ethics line, counts only with a school reference. A marker can reconstruct the study from this block alone.
Reusable share by week (Figure 1)
Four bars tell the story: 11.9 and 12.9 percent in the baseline weeks, then 26.8 and 31.4 once the discount began. The step lands exactly at week 3 and does not decay in week 4, which is the strongest pattern a before and after design can show.
Results
The results text carries the difference, 16.7 points with its interval of 14.5 to 18.9, the near constant sales volumes that rule out a customer mix explanation, and a costing: roughly 205 fewer single use cups per week at about 0.82 per cup avoided in claimed discounts.
Before and after by cafe (Figure 2)
The horizontal pairs show Library Cafe moving from 13.1 to 31.6 percent and Science Cafe from 11.7 to 26.5. Two sites moving the same way in the same week is the audit standing in for replication, and it is the second chart most single site student projects forget to include.
Discussion and conclusion
The discussion lists what moved together, discount, poster, weather and the end of exams, and why the step shape still favours the intervention. The conclusion recommends a term long trial with a control cafe and the poster separated from the discount, which is the claim the data actually licenses.
Conclusion points and references
Four bullets that survive a sceptical marker: the rise, both cafes moving together, causation not claimed, and the recommended trial. The three references are invented for this fictional study, because the reference block is required on this layout; on your board, replace them with your sources.
How to adapt this board
Put your before and after series into Figure 1 by week and your sites into Figure 2, and print raw counts beside every percentage so a marker can recompute them. Write the limits into the conclusion rather than a footnote; an undergraduate board that names its own weaknesses reads as stronger, not weaker.

What makes this board work

The claim is sized to the design

No causal verb appears anywhere on the board; the limits sit in the discussion and again in the conclusion points, ending with the recommendation for a controlled trial. That restraint is exactly what markers reward.

Counts and percentages both printed

298 of 2,410 and 713 of 2,450 appear beside 12.4 and 29.1 percent, so anyone can redo the division. A checkable board beats an impressive one in an examined setting.

Two charts, two questions

Figure 1 answers when the change happened, a step at week 3, and Figure 2 answers whether it happened everywhere, both cafes together. Neither chart is decoration; each carries one argument.

Questions people ask

What goes on an undergraduate research poster?

The same blocks as any research poster, abstract, methods, results, figures, discussion, references, with smaller numbers and honest limits. This board fills the standard framed classic layout with a two cafe audit, and nothing about the structure would change for a lab or survey project.

How much text should it carry?

This board runs about 780 words across its blocks, which reads in four to five minutes at a stand. The framed classic layout holds up to roughly 1,900 characters in each of the results and discussion columns; filling them matters more than filling every corner of the canvas.

Can I use a photo of my study site?

No poster layout has a fillable image slot, and this study does not need one: the two cafes are named in methods and the counting window is stated. If your project genuinely turns on appearance, put photographs behind the QR code and describe the measured feature on the board.

Should my supervisor be an author?

Usually yes, with their affiliation, and marked as supervisor if your department prefers it that way. This board lists the student as presenting author and the supervisor second; the author block generates the affiliation superscripts automatically, so adding another author costs nothing and loses no space.

What if my result is not statistically significant?

Report the difference with its interval and say what the interval rules out, exactly as you would a significant one. A null with a tight interval is a real finding. The pattern to copy from this board is printing counts beside percentages so the precision of your estimate is visible.

Which layout suits a formal department?

The framed classic used here, or the journal style board, both portrait and both conservative in structure. This one is on the prestige gold theme, and the theme can be swapped without touching content, so matching your department colours is a one click change, not a redesign.

Build your own in about a minute

The button below opens the generator with this use case already described. Change the wording to match your own, generate, then edit anything you like.

Make my undergraduate research poster

Other poster examples

Anthropology research poster example

Fourteen months of participant observation at a weekly produce market, laid out as an A0 portrait board in a journal style two column grid. The board treats ethnography the way a journal does: abstract, site, method, findings and discussion, with the counts a reader can check.

Psychology research poster

The strongest psychology research poster leads with its question and answers it in the first sentence of results. This A0 portrait example asks whether ten minutes of paced breathing before an exam lowers test anxiety, and answers with the number: 5.7 points lower with a 95% CI of -8.4 to -3.0, in a pre registered two arm experiment with 204 students analysed.

Linguistics research poster example

A vowel merger measured across 120 speakers in one coastal city, laid out as an A0 portrait board where a single hero figure takes the top half. The figure plots formant difference by birth decade, so the change reads as a trend rather than as two bars.

Want the steps in the builder? Read choose the template, theme and size, then change colours and fonts with themes. For everything this generator can do, see the poster maker.

Sources

Written and checked by the OneCraft team. Last checked .