Assertions about what is known should not be made lightly. They matter – especially when they are published in respected journals by respected authors. To the extent that poorly supported, or unsupported, claims are seen by readers to be credible – and practices are changed based on a belief that something is true that may not be – damage is done.
Last month, I wrote about what feels like a proliferation of poorly supported and unsupported assertions. (Note that I am not claiming to have studied this, just relating my sense.) In one post, I questioned a Harvard Business Review (HBR) cover article that, among other problems, drew conclusions about gender differences when it comes to conceptions of work-life balance. The analysis was drawn from a survey of 82 participants in an executive education class, 24 of whom were women (no joke). I shared my post by email and Twitter with both HBR and the Harvard Business School professor who was the lead author of the article and heard nothing back.
In a second post, I questioned this statement, made in the first sentence of a Stanford Social Innovation Review piece by Mark Kramer and Kate Tallant of FSG Social Impact Advisors: “Companies across industries—including education, technology, financial services, agriculture, and more—are playing a new role in education, one that improves learning outcomes while driving shareholder returns.”
In a comment on the SSIR site, I asked simply, “What data supports the claim made in the first sentence?”
This led to a back and forth in which SSIR editor Eric Nee chastised me for being a “naysayer” which he said was “not productive.” (His comment is remarkable for both its substance and its tone.) Steve Goldberg weighed in to argue that I was making a mountain of a molehill because the authors of the SSIR article had not said “all companies.”
But a number of others made clear, both on the SSIR site and on Twitter, that they supported my point – which is simply that claims about outcomes should not be made lightly. Supporting evidence should be provided. If it isn’t there, then write that companies across many industries are seeking to improve educational outcomes, not that they are.
Tris Lumley of New Philanthropy Capital wrote that “the importance of providing evidence to back up all claims cannot be overstated. Otherwise we are in danger of ‘shared value’ being a convenient concept for businesses that are just engaging in the equivalent of greenwash.”
Tallant, a co-author of the SSIR piece, whose initial response to my comment had danced around the question, weighed in again with this: “Outcomes do matter. Very much. And it’s true that most companies that are undertaking a shared value approach are still in the early stages of measuring the educational outcomes of these investments. For now, there are not a large number of examples.”
To me, this reads as a concession that the claim about outcomes cannot be supported (whether interpreted as Goldberg did or not) and, indeed, she went on to say she was “happy to edit the sentence.” It now reads “Companies across industries—including education, technology, financial services, agriculture, and more—are beginning to play a new role in education, one that improves learning outcomes while driving shareholder returns.” (Emphasis added.)
But this is only marginally better than the original, as it leaves in place the implication of a causal connection between corporate involvement and better outcomes.
Statements made about what leads to better outcomes matter.
We can do better.
Phil Buchanan is President of the Center for Effective Philanthropy. You can find him on Twitter @PhilCEP.