Analysis and conclusion
Processed data, graphs and a statistical result, all of it uninterpreted. Also, from step 3, a strategy and a tension that your conclusion is going to have to speak to.
- Every pattern named and explained
- An honest account of how far your data can be trusted
- A conclusion that answers the question in numbers
- A line connecting it back to the issue
This is the only place you are allowed to say what it means.
You have spent step 5 disciplining yourself not to interpret anything. Now do it all. Six marks, and they are not for describing what your graphs show; a reader can see that. They are for knowing why.
A strong analysis does three things, in order.
Name the pattern precisely. Which direction, how strong, and whether the statistics say it is real.
On its own, this is the lower band.
What process produced the pattern? This is where your abiotic measurements earn their place, because they are the mechanism.
This is what moves you up.
Take the answer back to the environmental issue and the strategy you wrote about in step 3.
This is where the work stays ESS.
Everything below is how we suggest you actually do it.
Explaining, not describing
A description says the line goes down. An explanation says what is pushing it down. The gap between those two sentences is most of this criterion.
Be precise about the relationship, too. A correlation is not the same as a difference, and neither is a trend. If you ran a test, say what the probability actually means rather than only that it was below 0.05. It is the chance of seeing a difference at least as large as yours if the two sites were really the same, so p = 0.0002 says that would happen about twice in ten thousand, which is why the null hypothesis gets rejected. What it does not tell you is how big the difference is, or that you have found the cause.
Your abiotic measurements do the heavy lifting here. They are the reason you can answer "why" at all, which is exactly why we asked for them back in step 4.
The managed piste scored higher for diversity than the forest, 3.23 against 2.18, despite having fewer species: six against ten. Backwards, on the face of it.
Simpson's index weighs evenness as well as richness. The forest floor is dominated by moss and bilberry, which between them take most of the cover, so ten species produce a low score. The piste's six are spread more evenly.
What causes the evenness? Canopy cover correlates with diversity at rho = −0.77 across all thirty quadrats, far more strongly than soil compaction at −0.41, recorded as penetration depth in centimetres, so a deeper reading means looser ground, and the sign only means anything once you have said that. Soil moisture shows nothing. Removing the canopy removes the shade that lets two shade-tolerant species take over.
That −0.77 needs one more check before it can carry any weight, and it is a check worth copying. Pooling two sites guarantees a strong correlation whenever the sites differ on both variables, so the number could be restating the site difference rather than explaining it. Run it inside each site separately and it survives: −0.55 on the piste alone (p = 0.036) and −0.66 in the forest alone (p = 0.008). The relationship holds where the site difference cannot account for it, which is what makes it an explanation rather than an echo.
When a result looks backwards, work through it in two moves: what your index is actually measuring, then which of your abiotic variables explains it. A result you can explain is worth more than one that came out as predicted, because explaining is what the marks are for.
Four things, one narrative
Bias, reliability, validity and uncertainty all have to be here. The common mistake is to give each one its own labelled paragraph, which makes them easy for an examiner to find and impossible to read.
Weave them instead. Take one trend, connect it to your question, then work through how far you trust it and why. It reads as thinking rather than as a checklist, and it sets up step 7 without you having to change gear.
The conclusion answers the question
It sounds obvious. It is the single most commonly forgotten thing in the whole report. A conclusion that says the topic is important, or that more research is needed, has not answered anything.
Restate your research question in different words, then answer it with numbers. Say whether your hypothesis held. And introduce nothing new: everything in the conclusion must already have appeared in the analysis above it.
If your data contradicts what you expected, say so plainly. An honest unexpected result is worth more than a tidy one you have leaned on.
Ski piste management has a significant effect on plant communities, and this is an important issue for the environment.
Simpson's diversity was higher on the managed piste (3.23) than in adjacent forest (2.18), a significant difference (t = 4.26, p = 0.0002), despite the forest holding ten species to the piste's six. The difference is driven by evenness rather than richness, and of the three conditions measured, canopy cover is the strongest candidate explanation.
Then travel back
Answering your research question finishes the science. It does not finish the ESS. Before you leave this section, take the answer back to the environmental issue you opened with and the strategy you argued about in step 3.
Two or three sentences is enough. What does your result mean for that decision? Does it support one of the perspectives you set out? This is the last place the two halves of your report can touch each other, and a report where they never do reads as a piece of fieldwork with an essay attached.
The plan proposes leaving some pistes ungroomed. On this evidence, that would let the canopy return, and with it the moss and bilberry dominance that produced the lower diversity score in the forest plots.
So the operator's claim is supported, but only on this measure. The result says nothing about whether an even mown sward is worth more than returning forest, which was precisely the tension in step 3. The data settles the number and leaves the argument exactly where it was.
Finish by taking your answer back to the strategy, and be precise about what it does and does not settle. Saying whose claim your data supports, and which part of the argument it cannot touch, is stronger than pretending fieldwork resolved a disagreement about values.
It can help you revise a general idea beforehand, what a p-value represents, how an index weighs evenness. That is ordinary learning, and it happens before you open this section rather than inside it.
It cannot interpret your results. Explaining why your pattern came out as it did is the six marks, and it rests on your site, your abiotic readings and your background reading, none of which a tool has.
Ready for step 7?
0 of 10The fifth one is the one people forget. Check it twice.
You have just written about bias, reliability, validity and uncertainty. Step 7 takes the same material and does something different with it: not how far your data can be trusted, but what you would change and what it would fix. The bridge is already built.
The ski piste investigation used throughout this guide is a teaching reconstruction. The location, sampling design, plant records and soil data come from real fieldwork at Mijoux in the French Jura. Canopy cover measurements were added to illustrate good practice and were not part of the original fieldwork. Photographs are the author’s own, taken at the site.