What is data literacy?
Understanding, analyzing, and communicating with data is a critical skill set for today’s data-driven world. Businesses make their most important decisions based on data, so having a deep comprehension of how data is organized, how to explore datasets, and how to convey that information to others is a highly-sought-after skill in the workplace.
Data literacy is about more than just selecting the right tool to slice and dice data. It’s the ability to read and interpret data, share insights through clear and compelling data storytelling, and make sound business decisions informed by analyses.
Below, we’ll share an overview of data literacy, why it matters, and several ways you can grow your data literacy.
Why is data literacy important?
Businesses depend on data to make strategic decisions. To make those decisions, they rely on data literate analysts and business intelligence leaders who can discover and translate all that complex data into actionable insights. As businesses generate and store increasing volumes of data, data literacy has fast become a critical skill for employers and employees.
A lack of data literacy can lead to strategic mistakes. Analysis results can be misinterpreted. And ineffective data storytelling can lead to poor engagement and slower decisionmaking. By becoming more data literate, individuals and teams can make decisions faster and more precisely, helping their organization stay ahead of the competition.
What are the essential skills of data literacy?
Becoming data literate is not easy, but it’s critical for thriving in the modern workplace. Here’s how you can increase your data literacy skills:
Develop a better understanding of your data
Before you begin any analysis, you need to have a deep understanding of your data and how it was collected. Is it qualitative (countable, or measurable), or quantitative (descriptive, or interpretable), or does your dataset contain both types of data? Is your dataset structured in a relational database, or is it unstructured data, which may require additional prep work to aggregate and format before beginning any analysis?
First, consider the source, quality, and completeness of your data:
How accurate or reliable was the data collection methodology?
Are there any null or missing values that might bias your analyses?
Examining data quality can save you headaches and valuable time later, as explorations or analysis of an unreliable dataset could lead you down the wrong path — or worse, prevent your team from making accurate decisions.
With your data formatted and prepped for analysis, where should you begin? Most of the time, analysis begins with exploring your data. The exploratory phase of data analysis is typically an unstructured journey through a data set, and most of the time includes some preliminary charts or other data visualizations to begin to make sense of the data. The goal is to conduct some initial discovery of your data to help you decide on the appropriate analysis.
When exploring your data, consider the following:
What types of variables are included?
Are there duplicates, outliers, or anomalies?
How are values distributed?
You can also build an exploratory data visualization through charts like histograms, scatter plots, and bar charts to help quickly answer these questions, so you can plan next steps for data wrangling and analysis
By asking the right questions and querying the data effectively, you can start to develop a hypothesis about what the data is pointing toward. Look for common trends, patterns, outliers, or anything else you can discover or glean. There are many exploratory data analysis tools that can help you quickly explore datasets, such as Observable Canvases, our collaborative data analysis platform, Looker, or Tableau.
Once you have an idea of what you’re looking to prove out, you can continue with your data analysis. Analysis can take many different forms, but may include anything from simple comparisons of categories to more complex statistical analysis or linear regression. The goal is to uncover any insight you can find that can help you better understand, or make decisions about, your business — and as you improve your analysis skills, you’ll develop a better sense of how to read and interpret data at the same time.
Bring data to life with data visualizations
With insights into the trends and patterns in your data, now it’s time to visualize what you’ve learned. Data visualizations are graphical representations or illustrations of information. They’re key to communicating the results of data analysis in a digestible and engaging way.
When visualizing data, consider which chart type makes the most sense for your data. Most data visualization software can quickly make common charts like bar, line, and scatter charts. But here are many, many different plot and chart types available, and a data-literate analyst is able to quickly identify the ideal chart or visualization that communicates its insights appropriately. Building familiarity with different chart types can help you add useful, underused chart options like beeswarm charts and horizon plots to your toolkit.
Beyond chart selection, there are a number of considerations that analysts and business intelligence leaders must think through when building visualizations. A well-designed data visualization communicates its insight clearly. It simplifies complex patterns, and provides helpful context in titles, labels, and annotations. In many ways, developing data visualizations is like telling a story: simplifying the plot, highlighting important insights, and directing attention to where it’s most needed.
It’s also important to consider accessibility when designing charts and dashboards. For example, Observable Plot includes accessibility features to make charts screen-readable, which helps viewers with visual impairments to understand a chart and the insights that it’s communicating.
Communicating effectively with data storytelling
Designing a data visualization is only part of communicating the insights gleaned from analysis — it’s important to know how to communicate your findings effectively.
Analysts should understand the audience of their analysis and visualizations, and what their level of technical expertise is. Does their audience know the terminology or common abbreviations used in their analysis? If not, a data dictionary might be helpful to better orient their stakeholders. It’s also important to relate the analysis back to broader business goals. For example, if an analysis of marketing campaigns showed that lead generation was up, how does that impact the overall go-to-market strategy of their team?
Instead of pointing to specific facts, craft a narrative with your data to help your audience remember your analysis — and act upon what they learned. Eliminate any extraneous detail that could obscure what you want your audience to understand. Explain what the numbers mean, and why the insights you gleaned matter. Share both context and takeaways to help your audience retain your analysis.
How can I improve my data literacy skills?
Here are some proven ways to grow your data literacy skills:
Practice makes perfect: Developing a high degree of data literacy involves lots of trial and error, but the more time spent with data, the better your data literacy skills. You can get started by exporting public data sets to data exploration tools like Observable Canvas, asking questions about the data, and attempting to answer the questions through careful analysis.
Share your analysis: One of the most important skills to develop in data literacy is learning to communicate and collaborate with others. Data analysis rarely happens in silos — so, as you’re developing your data literacy skills, share your analysis and explorations with your colleagues or peers. Ask whether or not your analysis is accurate, and if you’re communicating your insights clearly.
Find resources: There are also plenty of resources available to learn more about data analysis and upskill your data literacy chops. Look for online courses and books to start your journey to become more data literate, and look for examples, communities or meetups to get inspired or learn from data literate peers directly.
Toward a more data-literate world
Becoming more data literate is not just about honing technical skills or cutting your teeth in data analysis software. It’s about developing a curiosity about the world, thinking through and evaluating trends and patterns in data, and learning how to share your insights with others. There’s always more to learn, and sharing your analysis and perspectives with others can help cultivate your colleagues’ data literacy as well — helping your whole team move faster and more decisively.
Want to dive into a dataset to get started with expanding your understanding of data? Observable Canvases helps analysts and their stakeholders explore, analyze, and share insights within a flexible, collaborative whiteboard. Observable Canvases is currently in beta, but you can sign up for early access.