In a world where everyone and everything delivers data at every second, being data literate has become the key skill for outstanding in the working environment. Plenty of surveys are available showing how many organizations and people recognize their lack of ability to read, work, analyze and argue with data, but also their positive attitude in willingness to learn those skills.
Data literacy is a set of soft and hard skills that empowers people to understand data and informs them better to make database decisions. In addition to the main definition of data literacy per se there are other necessary partners like Data ethics or Data quality that complete the concept and are explored in this article.
A study by Qlik shows the state and impact of data literacy.
- Enterprise-wide data literacy is low: 24% of business decision makers surveyed are fully confident in their ability to read, work with, analyze, and argue with data. 32% of the C-suite is viewed as data literate, potentially holding senior leaders back from encouraging their workforces to use data to their advantage.
- Future employees are underprepared for data-driven workplaces: 21% of 16 to 24-year-olds are data literate, suggesting a gap in school and university educcation.
- Organizations are losing competitive advantage because better data literacy drives higher enterprise performance: 85% of data literate people estimate they are performing very well at work, compared to 54% of the wider workforce.
- Data is key to professional credibility: 94% of respondents using data in their current role agree data makes them more effective.
- There is an appetite to learn: 78% of business decision makers said they would be willing to invest more time and energy into improving their data skillsets.
In this article we are going to address various sides of data literacy. Through an in-depth exploration, we will delve into different perspectives surrounding data literacy, aiming to provide a holistic understanding of its significance in the contemporary world. This article aims to cover from different viewpoint multidimensions of data literacy, considering its impact on organizational dynamics, implications for decision-making, and its crucial role of evidence-based practices.
What is data literacy?
Data literacy is a skill that empowers people to understand data and informs them better to make database decisions. It helps to ask the right questions, build knowledge, make decisions, and communicate meaning to others. It’s not a binary concept but rather a spectrum of fluency that ranges from basic data-driven decision-making to more advanced data science, data engineering, and machine learning skills.
It spans across four abilities by definition:
The first two abilities enable us to define if what we see as data is a hoax or not before we find out.
- Reading data means being able to observe and comprehend it.
- Working with data means being comfortable with the information presented – not becoming good at coding or statistics.
The next two abilities enable us to move beyond observation to insights.
- Analyzing data describes the ability of being comfortable with asking questions, getting to the ‘why’ – making sense of the data.
- Arguing or communicating with data is the capability of interrogate the information as it is presented to you
It’s important to distinguish technical literacy from data literacy, something that organisations often ignore. Technical literacy covers the tools and related technical skills (e.g. coding, statistics) required for data science or analytics, while data literacy also alludes to soft skills like problem solving, decision-making and storytelling.
The Value of Data Literacy
Data literacy enables the adoption of a common data language throughout the organization that promotes mutual understanding. It is directly translated into more initiative-taking and better use of data. It is a key factor in the democratization use of data within teams and organizations by investing in tools and education that will help organizations go through turbulent times and anticipate and act quickly with confidence.
Key Components of Data Literacy
The definition of data literacy as the ability to read, work, analyze and communicate with data is just a part of the idea behind data literacy. Jordan Morrow, in his book, ‘Be data literate: The data Literacy Skills Everyone Needs to Succeed’ introduces his concept of the “data literacy umbrella”, which encompasses the following components:
- The data and analytical strategy lays out the roadmap defining the data literacy approach in an organization. It connects the needs to be fulfilled with the data literacy program and highlights its business value.
- With data science it’s possible to extract information and insights using scientific methods. From the author’s experience data scientists often lack communication skills. In this case data literacy is helping data scientists to create dialogue, language, and things everyone can understand.
- Data ethics describes the ethical principles behind how organizations gather, protect, and use data. The ability to understand and work with data, define its purpose and how it has been collected and measured, will directly determine the nuances of that understanding and our ultimate use of the data.
- Morrow defines data visualization as a simplified approach to studying data. Visual literacy helps connecting the dots and aids interpretation.
- Executive teams do not only own the data and the data and analytics strategy, but should also be the role models and enablers of data literacy initiatives. They are also responsible for aligning a data literacy strategy with the organizational goals.
- Data governance is the organization, process and policies how to gather, store, manage and dispose data. A lack of data literacy may result in collecting wrong, irrelevant or too little data.
- Data quality relates to how complete, accurate, relevant, timely, consistent, and trustworthy data is. If all those qualities are met, businesses are able to make better decisions. 100% accuracy is hard to achieve, but with a good understanding of issues, it may still be possible to base decisions on imperfect data.
Data Literacy Training and Development
The vast amounts of information generated every second can overwhelm any person or organization that wants to start working with data. Even though many can collect and store it, few are able to make benefit from it. Some organizations believe that the key is to empower their workforces to have the confidence to work with data efficiently.
This should not solely focus on individual employees, but also target the organization’s mindset. As any new process or learning skills takes time and effort, success should be built gradually and organically with a clear and invariable objective.
It requires setting up a common language, because a term can have different meanings depending on the point of view. A ‘customer’ could be defined as a website visitor by a data analyst and as a user who made a purchase by an accountant. For a common understanding of data it’s important to align on definitions across the entire organization.
In all cases the engagement of program leaders is essential to bring three elements together: culture, direction, and skillset.
Creating a culture of informed actions means stopping relying solely on senior management or experts to make a call, but using data-based evidence as basis for decision making and encouraging everybody in the organization to follow this principle. It is important that leaders empower the entire organization to learn about the organisations data, challenge analysis and generate insights for decision making.
There is no single skill alone describing data literacy – rather it is a set of skills. Some important hard skills like data transformation or data visualization should be fostered. Soft skills such as emotional intelligence, empathy, creativity, complex problem-solving, multi-disciplinary thinking and cognitive flexibility, should be addressed in data literacy programs. Programs that embrace human intelligence and augment it by technology will boost confidence and productivity.
Implementing Data Literacy in the Workplace
From our experience, there is no one-fits all approach to implementing data literacy programs as it depends on the company’s structure and complexity which approach is most likely to be effective.
Bottom-up and top-down techniques
The bottom-up approach’s goal is foster data literacy via empowerment (providing learning resources and encouragement) and motivation (emphasizing how the new skills and culture can make them more effective). This doesn’t mean that all employees turn into become data analyst or data, but enabling them to understand the value of their data and the importance of following certain standards or procedures when generating and analysing it.
A more challenging approach is the top-down approach focusing on strategy and processes that integrate analytics into daily team operations, effectively pushing the teams to work with data and become data literate.
For either approach, the following steps should be followed for successful implementation.
Identify and develop personas
Not all teams work in the same way with data or have the same goals associated with it, so it is important to identify those aspects and define different types of data personas. These can be identified using roles (data entry, data engineers, database administrator, data analyst, data scientist, data consumer, statistician) or from a businesss or objective perspective (data aristocrat, data knight, data dreamer, data doubter).
Finding the target
But how do you define who is the right persona to focus on? One option is by looking into the data literacy gaps for each persona and the potential impact of closing these gaps as well as how how much time/resource investment is required to close them. Keep in mind that a good target persona may also act as embassador in the organization to spread enthusiasm and skills.
The learning courses
Not all personas might need to close the same skill gaps. In a manufacturing company for example, an analyst working in an R&D department likely already possesses the necessary technical skills, but may need more training communicating data insights, whereas a production planner may need training on extracting and manipulating data.
Integration into daily work
Ensure that data literacy is integrated into the daily work processes. Encourage employees to apply their newfound skills directly to their tasks and projects. This practical application reinforces learning and demonstrates the immediate value of data literacy in the workplace.
Provide incentives
It could be a career path, a promotion, a bonus, or badges. Incorporating data for decision making should be incentivized instead of penalizing the opposite in order to encourage learning and provide psychological safety necessary for upskilling.
Measuring the success
The easiest way of measuring the success of your data program could be answering the question: has it helped you become more profitable? As employees’ data skills are growing and your organization has the power of making data-based decisions the benchmark will be surpassed. But that is not the only KPI that you can think about, because successful data programs should also result in higher employee engagement and motivation to take part in future data programs, and higher data quality.
The Role of AI Literacy
Nowadays organizations might ask themselves whether it’s more profitable for their business to invest in data literacy or AI literacy. With generative AI tools like ChatGPT or Stable Diffusion, AI literacy has never been more important. But how to approach AI literacy? Is it an extension, a complement of or something completely separate from data literacy?
Generative AI is influencing how enterprises change their approach towards data literacy. AI literacy implies understanding the data used for the training of AI models and how the models generate the outputs given this data and the input prompts. Biased outputs and hallucinations are a well-known problem, and understanding what hallucinations are, how to identify them, and why they occur are important aspects of AI literacy. Data literacy is closely related since the data literacy skillset can be applied to interpret if the outcome of a generative AI model is correct. The risk of poor data literacy in the context of generative AI is taking model outputs as facts.
Data literacy and AI literacy go hand-by-hand, because without a common language set by data literacy, a collective understanding of how AI projects are scoped, deployed and governed the context interpretation of the result of generative AI can lead to wrong decisions.
Data literacy is a key part of developing responsible AI, as it ensures diverse perspectives are incorporated into AI governance efforts, and that AI systems in production achieve better, robust and consistent outcomes.
On the other hand, generative AI may also support data literacy efforts and could be seen as a tool to make data more accessible and interpretable. AI assistants may lower barriers to insights by allowing users to query databases in natural language and visualize data without having to master visualization and dashboarding tools.
Conclusion
From a simple definition to a more deeper understanding of its components and how it applies at different levels from individual to organizations, becoming data literate involves developing a tailored data program that includes mastering both hard skills and soft skills and to be able to apply these in day-to-day data-based decisions.
Despite the growing impact of generative AI on tasks, data literacy remains crucial. It allows people to harness the intelligence of analytical data for decision making, a uniquely human capability that involves contextual understanding, judgment and reality checks in the interpretation of data.
References
- 2018 QlikTech International AB, “Lead with Data™ How to Drive Data Literacy in the Enterprise”, https://www.qlik.com/us/bi/-/media/08F37D711A58406E83BA8418EB1D58C9.ashx
- 2018 QlikTech International AB, „The Data Literacy Index”, https://www.qlik.com/us/-/media/files/resource-library/global-us/register/analyst-reports/ar-the-data-literacy-index-en.pdf?rev=988d7bbf8a1547878fca382a2f86dc2b
- 06.2020 DataCamp, Joyce Chiu, “4 Steps to Building a Successful Data Program”
- 2022 QlikTech International AB, “The Seven Principles of Data Literacy”
- 05.2023, DataCamp, Matt Crabtree, “Closing the Data Literacy Gap: Key Insights from the State of Data Literacy 2023 Report”
- Data Literacy project.org, 6-step Approach to launch a data Literacy Initiative
- DataCamp, Ted Kwartler, Haniyeh Mahmoudian, Sarah Khatry, Adel Nehme, “Data Literacy for Responsible AI”
- 2022 QlikTech International AB, “Data Literacy: The Upskilling Evolution”
- 05.06.2022, DLC, Luke Stanke, “The Essential Link between Data Literacy and Data Ethics”
- 03.03.2021, Kogan Page Publishers, Jordan Morrow, “Be Data Literate: The Data Literacy Skills Everyone Needs To Succeed”