guest post by Laura Sparks, Creative Sparks, for bbr marketing
“Data sources used in stories should be vetted for integrity and validity.”
Amen, and thank you to the good folks at Associated Press for including this new entry on data journalism in the 2016 edition of the AP Stylebook, released on June 1.
Data and statistics are the fuel that powers the content marketing machine. But misinterpreted and out-of-context statistics spread more confusion and hype than insights.
The 2016 AP Stylebook includes thought-provoking questions that all content creators and aggregators should ask to keep from sharing or reposting faulty data or incorrect conclusions. (For the full entry on data journalism and answers to all your AP Style questions, subscribe to APstylebook.com.)
The entry has some great pointers for firms that conduct original research, such as adjusting for inflation when comparing dollar amounts across multiple years. I have highlighted the questions that I find most useful for those of us who are typically reporting on others’ research, followed by my own musings on why we all need to ask these questions.
Does the data make intuitive sense?
Counterintuitive findings attract the most attention, so they’re favorites with headline writers and social media marketers. But review these statistics with a skeptical eye before passing them on down the line.
What is the original source for the data? How reliable is it?
It frustrates me when I click to the source for an interesting statistic and find that I have to go through one or more additional levels (and sometimes a separate Google search) before I find the original source. When reporting on others’ research, take the time to track down the actual report so you can verify the numbers and make sure you fully understand the context. Sometimes you won’t be able to access the research without a subscription. If that’s the case, and it’s not worth it to shell out the money, then I suggest looking elsewhere for data that is more freely available.
Why was the data collected? Was it for purposes of advocacy? Might that affect the data’s reliability or completeness?
Another of AP’s reminders is particularly relevant here: Correlations should not be treated as a causal relationship. Entities with an agenda could use correlations to draw conclusions that, once you dig down, aren’t supported by the data. There is no such thing as unbiased reporting. That doesn’t mean the research is flawed, but it does mean there is likely more to the story. Spend a little extra time to seek other statistics that complete the picture.
Is this the most current version of the data set? How often is the data updated?
I suspect more than a few content creators have given in to the temptation to use older statistics that are more in line with the point they are trying to make. Don’t do it. Find out if more current research is available. And if it doesn’t tell the story you set out to tell—use that as an opportunity to huddle with your subject matter experts to see how these new trends affect your firm’s point of view.
About the author: Laura Sparks crafts thought leadership content for accountants and other passionate professionals. Contact her at or 678-973-0687.