Using Data Visualization to Transform the Way You Work (5/5)
This is the fifth and final article in my data visualization series. This time, I show you how data viz can transform the way you work by using real-time, highly digestible data. It draws on all its predecessors in this series
In Let’s Use Florence Nightingale’s Secret Weapons, I talked about making data visual. You do this by using visual cues such as color, shape, and size instead of numbers and letters. And I convinced you (I hope) that when we do so, we make data much easier to understand and use. In the Data Visualization: Wielding Your Secret Weapons, I offered up tips on deciding what and how to visualize different types of data. In Sharpen Your Data Visualization Superpowers, I showed how to turn a good data visualization into a better one. Then, in How To Consume Data Viz Like a Superhero, I gave some pointers on quickly consuming vizes that you encounter in reports and articles.
Now let’s talk about using data viz to transform your work. We will start by considering how most of us work with data. Most organizations set goals, collect some information about those goals, look at data periodically or rarely, and then start furiously crunching numbers when a proposal or annual report is due. Perhaps we visualize some of this data to enliven the writing a bit and make us look more serious.
We don’t look at our data often because it’s difficult or unpleasant to do, and we are busy! Also, we might not think the numbers are a good reflection of what is actually happening. That's often because the data we have is dirty, disconnected, or just not appropriate for measuring our impact. (For more on the rational reasons we don’t use data, see Let’s Use Florence Nightingale’s Secret Weapons.)
Instead, we rely on anecdotes for our understanding of the impact we are having. We are particularly drawn to stories about individual clients or programs that confirm our expectations about impact. That’s because what we perceive is based not just on what we actually observe but also what we expect to observe.
This is how it works: The brain evaluates which of a variety of probable events are actually occurring. It also uses incoming signals from the outside world – what we might call data – to decide what it is perceiving. But there are far more signals coming from within the brain than data signals from the outside that affect our perception. These inner brain signals – or expectations – can distort our understanding of a situation. However, if we decide to consciously focus on data, we can perceive the world more accurately. That’s what the scientific method is all about: using data to confirm or negate our expectations.
Despite our aversion to data and our penchant for anecdotes, we invest significant amounts of time and money in data collection and storage. Many do it to comply with regulations or to satisfy funders. But fewer of us consider how we can put our data to better use if we make it more consumable.
Let’s consider two common ways organizations work with goals and data: annual reports and logic models. How can we transform them by making the data readily available and consumable?
Annual reports are, by nature, out of date. Sure, there’s a place for a summary of a whole year’s goals and progress toward them. But we also need to know what’s happening right now. Real-time reports are important, particularly for internal use. If you are tracking progress in real time, then you will know when you are getting off track and make midcourse corrections. Conversations about current trends lead us to consider factors that might be driving these trends.
Software like Tableau allows you to see your data in real time. You can even get Tableau for free if you are a smaller nonprofit (see more information on Tableau donations here). Or you can save your work to its server (see more information on Tableau Public here.) You can refresh the data by connecting your database and the application or through data uploads. And you can see your data from different angles using interactive filters. Check out this example and this example of interactive data dashboards.
The Tableau Foundation has a “living” rather than “annual” report. (They call it a “living annual report ,” which doesn’t make much sense to me.) It shows you everything you want to know about the grants they are making: who is receiving them, where grantees are, what grantees are doing. And it’s updated weekly. It also provides year-to-date and past-year views of their grantmaking. Filters allow viewers to slice and dice the information however they want. For example, by choosing “2014” and “United States” on their grantmaking map, you can see that they made grants in only three states that year. Check it out here.
Logic models are ubiquitous in evaluation and planning. A logic model (aka causal chain, model of change, roadmap, or theory of change) is a visual representation of how an intervention or program is supposed to work. It explains why your strategy is a good solution to the problem at hand.
Here is an example from the Greater Auburn Development Corporation in Chicago. It’s a logic model for a multiyear community education initiative.
There are at least four problems with many logic models:
- They are too simple. Some logic models are designed to take into account current conditions. Yet, most interventions/initiatives face unforeseen hurdles such as funding or staff changes. Logic models rarely take such instability into account.
- They are too ambitious. They assume that you can identify all the contributors to a problem and then influence them. In the real world, multiple factors relate to school drop out, homelessness, or obesity. And most organizations (or researchers for that matter) cannot identify them all. Nor can they fully understand how such factors relate to the problem they are trying to solve. Moreover, organizations often have little leverage over many of these factors.
- They do not make room for feedback and change. Organizations and collaborations learn like people do. They try out a plan, make mistakes, make adjustments, and try again. This cycle – which is key to “continuous quality improvement" – repeats on the road towards goals.
- They are not used. Countless meeting hours go into logic model development. Yet, they get more play in proposals than in daily work. If a logic model does not show the world you are facing each day, then it is not a useful tool. And it will gather dust in file folders and on servers.
By contrast, a living logic model can transform the way you plan and evaluate your work. Indeed it merges planning with evaluation. It helps you to assess your journey thus far and make route adjustments along the way. It may even cause you to reconsider goals.
A living logic model is more understandable and tangible than a traditional one. The user can scroll over any component in the model to learn more about it. Such descriptions can include photos and web links for interested users.
A living logic model shows progress to date. In the example below, color saturation indicates the status of each component. (See the interactive version here). Darker blues indicate more progress on the measure. And the user can click on any component to see what subgroups might be driving progress, stagnation, or regression. In the example, the growth mindset score measure is shown below the logic model. If you clicked on another measure in the model, say attendance, you would see a bar chart showing the school attendance for each child in the program. Clicking on a measure also brings up more information on the measure as shown in the second image below. Adding pictures and photos makes the program or initiative less abstract. Images show that the work is about real people involved in interesting programs and activities.
You can create a living logic model using free software like Tableau Public. And then use it every day to access real-time data to assess progress. This information, in turn, helps you to consider barriers and facilitators within and outside of the organization. And then you can make informed midcourse changes to the model and move forward on your adjusted route.
Living reports and logic models move data closer to your daily work. Data is no longer something that gets dusted off when writing proposals and creating annual reports. Instead, shows up at every staff and project meeting. It clarifies your work. And it guides your questions, leading to new and more effective strategies.
Editor's Note: This is Part 5 of a five-part series. Here are all five parts:
- Let’s Use Florence Nightingale’s Secret Weapons (Part 1)
- Data Visualization: Wielding Your Secret Weapons (Part 2)
- Sharpen Your Data Visualization Superpowers (Part 3)
- How To Consume Data Viz Like a Superhero (Part 4)
- Using Data Visualization to Transform the Way You Work (Part 5)