This past week I was able to attend a Management Concepts training offered by my Agency focused on presenting data analysis. This was a great three day training that if you ever have the opportunity to take I highly recommend. But as I know training dollars can be thin, I’m going to share some of my favorite take-aways and experiences from the class.
Define the objective clearly
This really has two perspectives. The first, if you’re the person asking someone to perform data analysis be specific about what questions you want the analyst to explore with the data. The second, if you’re the analyst use the specific questions to define your objective. Getting into the data can lead down rabbit holes with distractions of “eat me” and “drink me” and while these tangents may be fun to explore for us data geeks, they can eat valuable time and dilute your findings and implications.
Be Selective
In combing through the data you’re likely going to perform all types of analysis and end up with numerous results. There is a temptation to give someone all of it, to show those statistical muscles. Use the test you perform in your analysis to determine trends and decide what is going to be most influential in supporting the objective defined.
Keep visuals clean and concise
Likely you’re going to want to summarize some of your findings into charts and graphs. Excel has really improved over the years in allowing you to quickly and beautifully create these objects quickly. It is important to be conscious of your color choices. Red/Green color blindness is fairly common and so you may want to steer clear of those colors. Blues, Greys and Blacks can be good choices and using shade variation amongst them can give lots of options. Excel allows lots of customizations, but a good rule of thumb is to keep your charts and graphs in 2-D unless your displaying data in the 3rd dimension. Keeping it 2-D makes it easier for the reader to be sure of the value.
I don’t want to give away everything I learned over the week, but I would be curious to hear what tips and tricks you have for presenting analysis.
Opinions expressed are solely my own and do not express the views or opinions of my employer.
Sabrina Delay is part of the GovLoop Featured Blogger program, where we feature blog posts by government voices from all across the country (and world!). To see more Featured Blogger posts, click here.
Nice post, Sabrina, as usual. You offer some very helpful tips. My observations center on external data requests and dissemination…
It’s been my experience in interacting with the press and public over many years that a lot of folks who request government data either: 1) don’t understand the data being requested, 2) don’t comprehend an agency’s potential limitations to fulfill large data requests (unless the agency’s mission is data centered, like the Census Bureau or Bureau of Labor Statistics, for example), and 3) outsiders are prone to misusing and/or misinterpreting the data provided by agencies.
Therefore, it’s of critical importance for government employees who disseminate data to comprehensively explain and educate to the requester about what the data mean, what it doesn’t mean, and how to best use the data to draw accurate conclusions and make precise observations.
This is especially relevant for data requests and dissemination to/from the news media, academics, and key stakeholders. Just some food for thought. Thanks again, Sabrina, for the awesome post!
Edward Tufte’s The Visual Display of Quantitative Information outlines the Principles of Graphical Excellence, on page 51:
Graphical excellence is the well-designed presentation of interesting data–a matter of substance, of statistics, and of design.
Graphical excellence consists of complex ideas communicated with clarity, precision, and efficiency.
Graphical excellence is that which gives to the viewer the greatest number of ideas in the shortest time with the least ink in the smallest space.
These principles help guide every decision I make to either use or not use graphics in any of my products or presentations. Excel’s capabilities are far too often crutches for better design and analysis. Unless careful thought and time is placed on the product the result will be unreadable and mundane. I am glad to see the status quo is being challenged on data presentation in some levels of government. As more and more data becomes available how we process and present its contents is important in our role as public servants. Thank you for this article.