Week 6 · 2105620 Research Communication Skills for Chemical Engineers
10 September 2026
Today
Everything used today is on the sheet. Nothing needed to be brought.
24 September to 4 October · Weeks 1 to 6 only
Analysis of published work. Not recall of definitions.
Six capabilities, one per week
Every one of these is something you did in a room with your hands. None of them is a definition.
Block 2 · 20 minutes
The Results section reports what was observed. Everything else belongs somewhere else.
Could a machine with no knowledge
of chemistry have written this sentence
from the raw data?
If yes, it reports. If it needed a person to decide something, it interprets.
Words that smuggle a claim into the Results
| attributed to | Names a cause. Nothing was observed |
| indicating | An inference. Usually true, usually not measured here |
| confirming | The strongest verb available, and one micrograph does not earn it |
| excellent, superior, remarkable | Judgements. Excellent against what, reported by whom? |
None of these is dishonest. All of them belong two sections further on, where the reader knows to be sceptical.
Activity 1 · 20 minutes · five sentences on your sheet
Activity 1
All five read well. Four of them would pass a supervisor and a language check. That is the point.
10
Break · back at the time on the board
Block 3 · 20 minutes
None of these looks like a mistake. Every one of them is a default setting somebody failed to override.
In roughly the order you will meet them
| Truncated axis | A bar chart not starting at zero. A 2 % difference fills the panel |
| Dual axes | Two quantities, two scales chosen by hand. Any two curves can be made to agree |
| No uncertainty | No error bars, no n. The difference may be inside the scatter |
| Selective range | The measurement continued. The axis did not |
| False resolution | A smooth curve through four points, with a feature labelled between them |
| Image reuse | The same micrograph, rotated, in two panels. This one ends careers |
Same data, two axis choices


Nothing was fabricated. One number was changed: where the axis begins.
Activity 2 · 25 minutes · in pairs
Activity 2
Drawn for this exercise, in our own figure style, with invented data. Not from any published paper.
Debrief
| 1 | Axis starts at 96.5 %. A 2.6-point difference fills the panel |
| 2 | Twin axes, right one inverted, ranges chosen so the curves overlay |
| 3 | No error bars and no n. The difference sits inside the scatter |
| 4 | Measured to 900 cycles, plotted to 500. It fails at 520 |
| 5 | Four measured points, a smooth curve, an optimum labelled between two of them |
Four of these mislead using data that is entirely present. One withholds data that was collected.
Have you drawn any of
these five yourself?
I have. Twice.
Three things
Examination 24 September to 4 October, Weeks 1 to 6. Week 8 resumes on 8 October.
Before you go · 45 seconds
Five questions. Anonymous — no name, no email, no login.
The fourth question is the one I actually use.
The two most common answers open next week’s session.
oxidized-challenge-ed9.notion.site

Scan now — I will wait