Data Literacy for All: Simply Difficult

Never have charts and graphs been more prominent in the collective public consciousness. The increased focus on data-driven insights has, just as so much in life, been both positive and negative.

Readily available analytics and data visualizations provide common ground for discussion (aka debate) regarding critical issues—such as the reasons for evolving public health policies. They have also highlighted how little experience the general populace has with interpreting and applying such information properly.

A similar situation exists within enterprises as many intended beneficiaries of data products struggle to utilize them effectively—if they use them at all. Why? Today’s enterprise data literacy programs focus almost exclusively on enabling practitioners, from the business analyst to the vaunted data scientist. In this context, literacy is commonly understood as “developing competence and knowledge in a specified area.” Associated curriculums, therefore, emphasize the tools and methods to manipulate data and create analytic insights (from basic reporting to sophisticated AI algorithms).

Basic Data Literacy

Frequently missed is the most basic definition of data literacy: the ability to read and write and apply numeracy. Indeed, inherent in many data literacy programs is the expectation that every employee will have the opportunity and need to roll his or her own data. While a laudable goal, the work-a-day data intersection for most individuals is in the consumption, not creation, of insight.

It should not, therefore, come as a surprise that your largest data consumer segment has the most fundamental yet chronically undervalued and underfunded need: basic number sense. This may sound trivial or unnecessary—after all, doesn’t everyone know this stuff? Yet, recent public displays of numeric torture (even from those who should know better) belie this belief. And, as the saying goes, numbers lie—and even those with specialized training sometimes get tripped up.  

All of which is a long-winded way back to the primary point: If you want a truly data-enabled organization, everyone—and I do mean everyone—needs a grounding in the fundamentals of “reading data.” This is good for individuals and good for the enterprise, not to mention society at large. Done well, increasing basic data literacy will also create both the appetite and aptitude for putting analytics into action.

The Fundamentals

Not sure if your data literacy approach focuses enough on the fundamentals? Consider whether your employees can do the following:

  • Explain their own performance metrics—Can every employee clearly explain the intent and basis for their own performance measures? This should include describing the key data inputs, assumptions, dependencies on other people or processes, and variability in the results. Now, can they do the same for your enterprise KPIs?
  • Apply insights in the context of their work—Does every individual understand the context, intended use, and outcomes of the analytics insights they see? Does your customer service agent, for example, understand when and why to choose one recommended service over another? Can a machine operator confidently decide when—and when not—to override an algorithmically determined setting? And when to act or ignore an alert when it sounds?
  • Interrogate a new data visualization, metric or recommendation—Yes, interrogate. Individuals should be able to accurately interpret the information presented as a metric, chart, or graph. They must also be trained to identify the source and assess the context in which information was gathered in order to apply it properly. As well as when to….
  • Ask for more information—Have you armed your employees with the language to clearly articulate their concerns? Are questions even encouraged? When questions arise (as they should), is it easy to connect with the data or analytics experts who can respond in a timely fashion and layperson’s terms?
  • Identify new opportunities to use data to improve their work—Have you trained individuals to spot processes, decisions, or tasks that could benefit from analytics or data? Do they have a rubric for translating common questions or situations that are ripe for analytics into common analytics approaches?

If every employee, regardless of the role, cannot confidently do the above, your data literacy program may be losing sight of the forest for the trees.

Photo by Campaign Creators on Unsplash


Subscribe to Big Data Quarterly E-Edition