/ Experimental Data

Tips for presenting your research data

Let me set a scene for you. You've packed your bag, signed in, and donned your name tag. Coffee in hand, you are ready for the first talk of the day. And then it happens. The presenter pulls up their first graph with a barely legible axis and says, “I’m sorry if you can’t see this clearly.”

Unfortunately, this is rarely a one-time occurrence at a conference. As scientists, we sometimes spend so much time fixated on our research that we either can’t see possible sources of confusion or simply don’t take the time to make clear figures.

Don’t be that scientist!

Logistics

For the average lab that isn't yet cloud connected, your data-producing pieces of equipment are tethered to a computer. Between equipment turnover and new purchases, you're likely also dealing with a fair amount of dated PCs. This means trying to get data off of computers that have limited USB ports or even worse, the barely-holding-it-together, must-get-just-the-right-angle-to-register-your-flash-drive type of port. You know the one. Additional challenges come up when computers aren’t online or have disabled USB ports for security purposes.

TIP: Enable easy access for your entire lab by connecting instruments to the cloud. Bonus? You get remote access for working from home on a snowy day, or grabbing information in real time when you're actually at the conference.

Quality

Your figures always look great on a small laptop screen. It's the transition to a much larger presentation screen that trips people up. A common culprit is the microscope image.

TIP: Use a high-resolution tif to get beautiful, crisp figures. Also keep in mind that light balance for fluorescent images can be tricky. Depending on the quality of the projector, images can easily appear way too dark.

I once gave a presentation in which two side-by-side projectors were set up, but they weren't the same quality. Half the audience was treated to black boxes with subtle hints of color while the other half saw gloriously colored confocal images.

TIP: If at all possible, preview your images on the projector. Conferences often set them up so you can flip through your slides before presentations start, or during a break.

Images can come with a data zone on the bottom detailing date, a scale bar and some other information. While useful to the scientist, they don’t tend to fare well when magnified. Especially if images are assembled together, the multiple data zones get very small. The result is a nice clear image with some fuzzy text on the bottom and a hard to read scale bar.

TIP: It’s an easy fix to crop around your nice image and add a scale bar separately. ImageJ is an easy-to-use, free software option that also gives you flexibility for scale bar color and location.

Presenting your research data images | TetraScience Blog

Consistency Issues

When presenting, colors and patterns are easy ways to designate different experimental groups. It allows the audience to follow your presentation with ease.

TIP: To be effective, be sure to carry out your colors and patterns through the entirety of your presentation - from the introductory diagrams through to the supporting graphs.

It is relatively straightforward when you're working from a single program like Excel, Prism, or Matlab. But start switching between multiple programs and you face a host of challenges. Do your best to keep all of your figures standardized in color scheme and style regardless of the software the data comes from.

TIP: Make sure your axes are the same size, color and weight. Keep all your fonts the same size and style, and most importantly, be sure that they are large enough to be easily read.

Don’t be content with the default excel graph. Everyone has seen the default and it can give the impression that you don’t care. The automatic lines around the graph look especially clumsy when you are presenting a number of graphs and images together.

TIP: Delete the background grid lines. Change the axis to black, customize the scale, and change the font style and size. Move the legend to the upper left so it is visible but doesn’t take up a lot of space. I also change the colors of the column bars and use capped error bars.

Compare the default saved as a jpeg to a cleaned up high resolution tif!

Presenting your research data graphs | TetraScience Blog

When working with an Excel graph, I often import into Gimp with a high dpi (300-600) and export as a tif. This is much better quality than saving as a jpeg directly from excel which is shown below on the left. There are a number of similar programs, but Gimp is free and straightforward.

Presenting your research data graphs | TetraScience Blog

Working with Specialized Software

At least Excel and Prism are designed to make figures. Keeping figures clean and consistent becomes more difficult when working with multiple programs, where analysis is the goal and a figure is a byproduct.

In many cases raw data can be imported into Excel or Prism, but not always. Chemists have NMR and mass spectra. Material scientists have DSC readings, and cell biologists have flow scatter plots. Not all of the data-specific software make figures easily or well.

TIP: It is difficult, but try to keep your colors the same or at least similar. Don’t be afraid to play with different formats to get the cleanest, highest quality image.

Videos

Having videos embedded in your presentation is often an affective way to showcase your research.

I've got some bad news though. It will fail on you. It will work perfectly at home. It will work just fine in the office. It will even work for the practice talk. But during the real thing, it will fail.

TIP: Test your video multiple times. If they force you to transfer your presentation to a laptop with a flashdrive, then try to find out beforehand what version of powerpoint they have. I've watched more than a few mismatch of powerpoint versions cause a video fail.

Now you know all of my presentation tricks, so how about you share a few? What are your go-to tricks for building a presentation that showcases data? Leave them in the comments below!

Kristie Tevis

Kristie Tevis

Kristie holds a PhD for Biomedical Engineering from Boston University and has over a decade of research experience. She is currently working in protocol development and IP data organization.

Read More