real life, Writing

A Thousand Words: Figures for Articles and Presentations

Graphs and pictures are extremely important parts of every lab report, journal article, and presentation.  However, not every scientist has the ability to tell the story properly.  If one cannot tell at a glance what it means, or even on longer viewing is confused, the graph or picture is bad and should be clarified.  Worse, unethical scientists can easily use these to lie to and mislead their audience, though a basic literacy can help prevent one from falling for these deceptive figures.

For one good example, go back to my Correlation vs Causation post.  Just because two things happen at the same time, or even have a very similar graph when plotted, does not mean that one causes the other.  For instance, just because the sun rises every day you’re alive does not mean that the sun rising causes you to continue living (or vice versa), because the two aren’t related (unless you go the distance of saying that without the sun, you probably wouldn’t be living, in which case…that’s true, but a bit of a stretch).  It’s easy to notice when it’s something like cheese consumption and number of anime getting dubbed in English and less so when it’s, say, giving a new drug to people and them getting better.  Preventing this misconception during an experiment involves, on the part of the scientist, eliminating any extraneous variables possible (such as trying to make the control group as similar to the experimental group as possible).  The clues to discover this on the part of the reader usually can be found in the rest of the report, article, or presentation.  The burden of proof is, unlike in American law, on the scientist, rather than the scientists’ critic.  (Conversely, if others try to replicate the experiment and cannot achieve anything resembling the same results, the validity of the experiment is called into question.)

The title of the figure should also be examined.  If it does not accurately describe the contents, it’s likely that the figure was changed last minute and the writer didn’t catch that change (that, or the writer is trying to pull a fast one).  Similarly, the axis labels (x being the one on the bottom and y being the one on the side) should also be read if it’s a graph, as it should tell you what two variables the scientist is relating (and is subject to the same scrutiny as the title, with the added check of correlation vs causation).  For example, does it make sense for, say, the height of plants to be related to the amount of a newly developed pesticide applied to said plants?

Another important thing to keep an eye on is the scale of a graph.  Some really weird things can happen if you mess with the scale of the graph.  Mostly, they can make mountains out of molehills, quite literally (which is fine if you want to examine the molehill in great close-up detail and make it obvious to the reader what is happening), or make a quite drastic change appear flat.  It’s most important to see if the graph starts at 0 and the difference between points.  (If no scale is present, it’s a bad graph or a bad scientist, either because they’re incompetent or a used car salesman.  If you know what you’re doing, you show the scale on the side and you make it obvious by pointing it out in the presentation, text, or caption itself.)

Legends are also incredibly important.  If it’s a line graph, you absolutely need to know what every color on the graph means.  If things are being compared, in order to read the graph correctly you need to understand what the things being compared are.  The legend is essentially the cheat sheet explaining important parts of the chart according to the author, so carefully reading these can be especially important.

Pictures are good for depicting equipment, lab setup (particularly when elaborate—I’ve seen some setups that resemble Rube Goldberg machines), or qualitative results, such as the color of the resulting solution.  Does the picture cover one of these categories or otherwise communicate information that you wouldn’t otherwise know, or is it just added to look pretty?  (Nothing wrong with one or two of those; it’s helpful to get people’s attention.  If that’s the only thing the pictures are adding, though….)

As mentioned in the opening paragraph, the figure should be readable at a glance, or at least without too much time and effort.  If a reader who previously understood what was happening suddenly is confused, you’ve made a mistake.  The figure is meant to contribute to, not detract from, the rest of the text.  (In a presentation, clarity is even more important, because you want people to be examining the pictures in your powerpoint or looking at your poster and listening to your words rather than just reading ahead and being bored.)

In general, your reading of the graph should be “what is this graph trying to say” and “what is it actually saying”.  Despite the fact that I said when designing graphs you should attempt to make it fast and easy, you shouldn’t assume that all graph makers are keeping this advice in mind.  Keep your critical thinking skills alive.

Do a quick search on ‘misleading graphs’, and comment about your favorite!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s