Archived Webinar: Presenting Data Effectively

Crystal clear graphs, slides, and reports are valuable – they save an audience’s mental energies, keep a reader engaged, and make you look smart. This webinar held on June 6, 2017, covers the science behind presenting data effectively and will leave viewers with direct, pointed changes that can be immediately administered to significantly increase impact. Guest Stephanie Evergreen also addresses principles of data visualization, report, and slideshow design that support legibility, comprehension, and stick our information in our audience’s brains.Evergreen’s presentation was followed by an audience question-and-answer session, which is included in the recording. Not all the questions were answered at the time, and Evergreen answers some additional session questions below.https://youtu.be/8BGJjYJisyAEvergreen is an internationally recognized speaker, designer, and researcher best known for bringing a research-based approach to better communicate through more effective graphs, slides, and reports. She holds a PhD from Western Michigan University in interdisciplinary evaluation, which included a dissertation on the extent of graphic design use in written research reporting. Evergreen has trained researchers worldwide through keynote presentations and workshops, for clients including Time, Verizon, Head Start, American Institutes for Research, Rockefeller Foundation, Brookings Institute, and the United Nations. She is the 2015 recipient of the American Evaluation Association’s Guttentag award, given for notable accomplishments early in a career.She is co-editor and co-author of two issues of New Directions for Evaluation on data visualization. She writes a popular blog on data presentation at StephanieEvergreen.com. Her books SAGE Publishing books Presenting Data Effectively and Effective Data Visualization both reached No. 1 on Amazon bestseller lists. A second edition of Presenting Data Effectively was published in May.

***

  1. When is it best to place the data information (e.g. 20 percent) on a bar or lollipop vs. using a scale on the side or bottom of a chart?

If people will want to know the exact value, add the data label. If the overall pattern of the data and estimated values are sufficient, use a scale. But don’t use both – that’s redundant.

  1. How do your clients and colleagues respond to the ‘flipped report,’ in which research findings and conclusions are presented before the discussion, literature, methodology, and background sections?

With a “duh” as in “Why haven’t I thought of that before”? Generally, clients appreciate how a flipped report values their time. On occasion, you and I will find audiences who really bristle at the idea, usually people steeped in the academic culture, so check first if a flipped report structure would be okay.

  1. Any tips for the converted about changing resistant organizational culture to data visualization? "You need to use our template!"

Culture change is slow, so the first tip is to be patient. After that, try remaking one of your own old (bad) slides or graphs to show what an overall would look like. See if you can get a friendly client or customer you know to give you feedback on it. Then report on the redesign and the feedback to others in your organization. Try getting someone from senior management on board. Leave a copy of my book in their mailbox or in the break room. And hang in there.

  1. How do we report small numbers? Without percentages?

I would report small numbers as raw numbers, not percentages. Try an icon array for a visual.

  1. Where is the best place to get report templates?

In your imagination! Any report template is going to look like a report template, not like something that fits your own work. Look around for inspiration, for sure, like on my Pinterest boards, but create your own style that fits you and your work.

  1. What program do you use to create dashboards or infographics? We've used Piktocharts.... are there others?

I work within the Microsoft Office suite. I make dashboards in Excel and infographics in PowerPoint. This way I have total control over the design and everyone on my team can make edits. A quick Google search of either dashboard or infographic programs will give you hundreds of choices you could work with. If you want something from that list, look for maximum flexibility, low learning curve, and reasonable expense.

  1. Each chart can have multiple findings; are we skewing the results when we highlight certain findings over others using color and data?

“Skewing” sounds like we are manipulating, but that’s not the case. Using color to highlight a certain part of the graph still leaves the rest of the graph completely intact and able to be seen. Adding color does, however, reflect an interpretation we have made of the data. But that isn’t “skewing” – it’s telling people our point and that’s why they are listening to us in the first place.

  1. Can you please explain the difference between your two books? Thanks!

Sure! Effective Data Visualization walks you through how to choose the right chart type and then how to make it in Excel. Presenting Data Effectively talks about formatting graphs well with consideration of text and color and broadens that discussion to address dashboards, slides, handouts, and reports.

  1. One challenge I face is presenting nuanced findings in an accessible way. For example, when there are limitations to the data or subgroups that need to be acknowledged or findings need to be interpreted with caution. As a researcher, it worries me that the client might put tentative findings "out there", misrepresenting them (to a degree).

This makes your title and subtitle ever more important. Be very clear in your wording that the findings are limited. You can also add things like confidence intervals to your graph if you are confident that the reader will know how to interpret them. If it is still going to be a concern, don’t make a graph of the data. People are drawn to graphs because we look at pictures so don’t put the data in a picture if you are worried people won’t read the nuanced narrative. 

Previous
Previous

A Cure for Cancer, More Likely with Data Sharing

Next
Next

How Industry Can Help Advance Computational Social Science