Data slides (3)

ffolliet February 20, 2017 3 Comments

Data slides must be easily interpreted, memorable and clear. They will support the message, not distract, cannot be taken directly from a scientific document and are best if their theme, font colour fit with the other slides in the presentation.This is good design. It strongly influences audience engagement, interpretation and retention of the message. Good data slides are essential. They are not about providing a massive data download but a message. The data is in the scientific paper.

This blog is not a discourse on design principles and one need not wear roll neck shirts, drink organic Guatemalan espresso and use only Mac products to design a great p2. Application of simple design principles to illustrate a message will significantly improve a presentation.

Continuing to use the presentation on Crash-2 and this delivery

“This slide showing the relative age bands in the trial comparing the study and the placebo group. You can see that for each age band the numbers are very similar, under 25, 2783 received tranexamic acid and 2855 were in the control group, 3012 aged between 25-34 received the drug and were compared to 3081 in the control. There was no statistical difference between these age bands or in fact any age band.”

This might be improved by application of some of the following principles:

1. Improve signal to noise ratio
2. Use simplicity
3. Use colour
4. Use simple explanation
5. Image superiority
   1. Improve signal to noise ratio
   It’s like tuning a radio, if anyone does that anymore. Maximising the impact of the message by reducing the clutter around it. A data slide needs no additional clutter such as background images, reference lines, multiple data labels gradient colour fills, University logos, template background, references and the like, Remove any of these until only the single message of that slide is left. The slide cannot stand by itself, it is part of a presentation, complete only with the story and its delivery. It is not a handout image.2. Use simplicity
   This slide contains a lot of data. That data may be important for the presentation as a whole but this is best stripped down so that the audience can concentrate  on the particular section of the topic being discussed. It is important to remember that the presentation must not turn into either a massive data download or a show containing hundreds of slides. try to reduce the complexity of the message. Percentages are important but in this image they do not add to the specific message at this time.
   3. Use colour
   Colour adds interest. It may also add glare and distraction. A palette of colour ties a presentation together and should be chosen at the end of design to unite and fit with the whole of the presentation. It should be used with restraint and facilitate understanding. The yellow above is completely washed out and together the four colours clash. Use colour to highlight a particular group, perhaps the study arm.4. Use simple explanation
   As highlighted earlier, a data slide must be interpretable promptly, ideally within three seconds. An explanation, rather than the title of the table within a document, may give clarity. The purpose of the data slide is not to deliver data but to support the message being delivered. No one in the audience will remember details of the 4×2 table, they will remember the message it has. Make this clear. The data table will be in the handout/ original paper.
   
   5. Image superiority
   Finally, images are superior in conveying a message compared to numbers. Can you recall the message regarding acute intra-cranial haemorrhage from the previous post? Bare numbers remain simply that. Their import comes when related to a message. The message in this section of the presentation can be trimmed significantly from the attempted data download above. Not the repetition of style and font and colour.

   “The precise numbers are in the handout. The essential fact is that there was no statistically significant difference between the study groups in any of the age bands.”

Simple design principles can significantly improve data slides. Ensure that a slide has a single message and deliver that clearly. The message will then be retained. The audience will never remember the individual numbers in the different arms of the study groups, but you know the import of the statistical difference between the two groups. That’s how to deliver data.