Literacy rates are commonly tracked across the globe. A foreseeable indicator to be tracked in the future as we head further and further into a data driven world will be data literacy rates. While harder to define and track, understanding these key 12 guidelines for spotting bad data analysis would be included in my definition of being data literate.
Starting with sensationalized headlines (how many links have you clicked on that started similar to the title of this post?), these 12 points cover basic issues with data claims that we all will need to be aware of as data increasingly blankets our day to day, such as knowing the difference between correlation and causation.
On the flip side, it’s increasingly important that media know how to how to interpret data well. Not only because it is a valuable tool to tell a story, but so that they can avoid making those same 12 data analysis pitfalls. More and more journalism is moving into data visualization and analysis, as outlined in this post about the media in Africa taking further advantage of the increasing amount of open data.
While important for data analyzers and data readers, it’s imperative to start off with good, quality data. ODI has launched a report about how the data available to development stakeholders, and how it is insufficient for the needs of a world that seeks to eradicate poverty, and the ways we might use new uses and forms of data to reach our goal.
90% of all data readers believe the majority of statistics they read, I figure we should all second-guess (since I made it up), and a topic to truly consider as we seek to make data more available to the masses.