18  Presentations

This module is assessed differently from the others. Instead of a test, you give a short presentation in a lab session. That is deliberate: the point of the module is communicating to a person, and a written test cannot check whether you can do that.

Most of the advice from Chapter 17 still applies – lead with the answer, one chart one point, say what you do not know. What changes is that your audience cannot re-read anything. They get one pass, at your pace, and whatever they missed is gone.

Learning Objectives

By the end of this chapter, you will be able to:

  • Structure a short talk so the finding arrives early
  • Build slides that support what you are saying rather than duplicate it
  • Present a chart out loud, at the pace a listener needs
  • Handle a question you do not know the answer to

18.1 One Pass Only

A reader who loses the thread turns back a page. A listener who loses the thread is gone for the rest of the talk, and you will not know it happened.

Everything else in this chapter follows from that. Say the important thing more than once. Put fewer things on a slide than you would in a report. Move slower than feels natural, especially through anything with numbers on it.

18.2 Structure

For a short talk – five or ten minutes – this works:

  1. The question, in one sentence. What were you trying to find out, and why would anyone care?
  2. The answer, in one sentence. Do not save it. A listener who knows where you are going can follow the middle; one who does not is just waiting.
  3. The evidence. Two or three charts, each with one point. This is most of your time.
  4. What you are unsure about. Brief and honest.
  5. The answer again. The last thing you say is the thing they remember, so make it the finding rather than “any questions?”.

The difference from a written report is item 5. In a report the reader can flip back to the summary; in a talk, the ending is the summary.

18.3 Slides

A slide is there to support what you are saying. If it says the same thing you are saying, your audience reads it instead of listening, and reading is faster – so they finish before you do and then wait, or worse, get ahead of you and stop paying attention.

What this means in practice:

  • A headline, not a label. “Canola yields 15% higher in the black soil zone” tells the audience the point. “Yield by Soil Zone” makes them work it out from the chart.
  • One idea per slide. If you need to say “and also” you probably need another slide.
  • Few words. Enough to anchor what you are saying, not a script.
  • Big enough to read from the back. Which is smaller than you think – check by standing up and walking away from your own laptop.
  • Charts need to be bigger than in a report. A figure that reads fine on a page is often unreadable projected. Fewer series, larger text, no fine detail.

Fifteen slides for a ten-minute talk is too many. Six or seven is usually about right.

18.4 Talking Through a Chart

This is the part most people rush, and it is the part your audience most needs you to slow down for. You have been looking at your chart for hours. They have been looking at it for four seconds.

Say, in order:

  1. What the axes are. “Yield in bushels per acre going up, year across the bottom.”
  2. What one mark means. “Each point is one risk zone.”
  3. What to look at. “The thing to notice is the gap between these two lines after 2022.”
  4. What it means. “So the black soil zone pulled ahead once the drought ended.”

That takes about fifteen seconds and it is the difference between an audience that follows your argument and one that is still working out what the axes are while you deliver the punchline.

18.5 Questions

You will be asked something you cannot answer. This is normal and it is not a failure.

The good response is short: say you do not know, say what it would take to find out, and move on. “I did not check that – I would need to look at whether the zones differ in seeding date as well.” That is a perfectly respectable answer, and it is a considerably better one than guessing, because a guess that sounds confident is the same failure mode we spent Module 5 on.

Two other things that help. If you do not understand a question, ask for it again rather than answering the question you wish you had been asked. And if someone finds a real problem with your analysis, say so plainly – being visibly willing to be wrong is worth more to your credibility than the point you lost.

18.6 What Is Being Assessed

The presentation for this module is short and low-stakes. I am looking for whether:

  • the question and the answer are clear within the first minute;
  • your charts are readable and you explain them rather than displaying them;
  • the numbers you quote are ones you can account for;
  • you are honest about what the analysis does not show.

Polish is not on that list. Nervous delivery is not on that list. The one thing that does not survive is a talk where the audience cannot tell what you found or why they should believe it.

  • Michael Alley, The Craft of Scientific Presentations – the assertion-evidence approach: every slide headline makes a claim and the visual supports it. Directly applicable to the headline advice above.
  • Cole Nussbaumer Knaflic, storytellingwithdata.com – the makeover posts are as useful for slides as for charts.