When denial of data becomes dangerous

There was a great story on Moyers and Company on NPR last night — “The Toxic Politics of Science” — which left me thinking not just about the chemicals that we live amongst, but also about confirmation bias, again, and about how data is suppressed, denied, and even used to deceive.

On the episode, David Rosner and Gerald Markowitz, authors of Lead Wars: The Politics of Science and the Fate of America’s Children (California/Milbank Books on Health and the Public), discuss the hazards of lead, the ethical, legal, and scientific history of the use of lead, and the industry that worked to deny and discredit scientific research that demonstrates the dangers of lead for many decades.

On the show, and I assume in their book, lead poisoning was also compared to other chemicals and the industries that have worked to deny and discredit scientific research indicating that those chemicals are dangerous. Think BPA (Bisphenol A) but also asbestos, and many others. They speak about how industry has mounted efforts to discredit good research, and develop their own, less credible research (and therefore data) about each of these chemicals. The scariest part? Many of these chemicals not only still lurk in our homes, but there is little responsibility (if any) taken by the industries that often knew these things were dangerous long before the public understood so.

While most of the data I work with is not nearly so ‘high stakes’ as that to do with chemicals like these, I do see parallels between the tactics of industry and the tendencies of regular folks to reject displeasing findings. Fortunately, there are not such dark forces at work, and simpler ways to encourage visibility and use of data from lower stakes work — no lawsuits required here. But just as one solution to the chemical problem may be better regulation and accountability for industry, one way that we, as a society, encourage transparency and use of findings in social services, education, and the like is through accreditation or evaluation requirements. A stick, rather than a carrot, certainly. I’d rather work proactively to learn from attend to findings (to actionable data), but sometimes it seems a stick is warranted.

Advertisements

3 thoughts on “When denial of data becomes dangerous

  1. Sheila B Robinson, Ed. D

    Methinks you’re looking for info to confirm your confirmation bias! Seriously, you make excellent connections here and they remind me of what I read in the insightful book Proofiness: The Dark Arts of Mathematical Deception (if you’ll permit the link, I wrote about that book here: http://sheilabrobinson.wordpress.com/2013/01/28/they-say-that-…bers-never-lie/). Learning about the endless possibilities of deception with data has changed the way I read, think, and live in the world. It’s scary to think about just how much is suppressed, denied, or manipulated. We continue to be encouraged to use data to inform our practices, but we’ve barely touched on the conversation about which data, and whose data to use and why.

    Reply
    1. Kim Post author

      Yes! I’d forgotten about that post of yours but should totally have noted it here, because it’s totally apropos (my favorite term of the week, though I’m probably overusing it, but that’s another topic entirely). Thanks for making that connection!
      And yes — there’s a call to action here, I suppose, to have bigger conversations about “which data, and whose data to use, and why” as you so smartly note.

      Reply
    2. Kim Post author

      Also, Ha! I probably am operating with some confirmation bias about the prevalence of confirmation bias here too. How meta. 😉

      Reply

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s