The rigor of representation

You wouldn’t claim to have identified a meaningful pattern in data if you had a p-value of 0.6. Gender is random at birth, but you wouldn’t try to publish a randomized trial if your randomization strategy involved putting boys in the intervention arm and girls in the control arm. And you wouldn’t try to attack the individual mandate in a universal health insurance program without specifying how to avoid the death spiral.


All areas of public health have basic standards of rigor—norms that determine whether what we’re saying is meaningful, and that ensure that we’re careful about what we say and how we say it. While it is common to recognize, practice, and teach rigor in statistical statements, for example, or economics, it is uncommon to even recognize rigor in visual statements. Yet if a picture is worth a thousand words, it’s important to be a thousand times as careful in what we depict visually as in what we describe verbally.



Screenshot 2016-04-22 11.21.30

Visual mistakes may seem trivial because visual culture seems fluffy. But they’re not and it isn’t. Visual mistakes are real and can be terribly damaging. Visual mistakes can alienate people we’re working with, but because we have so little common language to talk about visual rigor, people who are harmed by visual mistakes have little recourse. It isn’t as if they can jump up and say, “Wait a minute! Don’t you know what John Berger says about this?” So while the critique of statistical mistakes is mediated through more or less genteel article reviews, the critique of visual mistakes comes out as sarcasm and mockery. But it’s sarcasm and mockery that should be taken seriously because the visual mistakes matter just as much as the statistical mistakes.


Lauren Kascak and Sayantani DasGupta have a thoughtful piece in Sociological Images about the role of photography in global voluntourism.

In the end, the Africa we voluntourists photograph isn’t a real place at all. It is an imaginary geography whose landscapes are forged by colonialism, as well as a good deal of narcissism.

A thoughtful insight, and one I would extend to academics as well.  For them, I don’t think it’s narcissism at play as much as sloppiness among people who aren’t used to being disciplined about visual culture.


a visual mistake
This kids are orphans. Or is that a Visual Mistake? Read the Onion for the truth!


We’ve all seen images like these that are posted without irony. What did their authors mean to say and what did they actually say?


There’s a lot to mock in these images. Beyond the mockery, there are a lot of visual mistakes to critique. They are usually static, with no work being done; the people are often either in states of complete desperation or complete relaxation; there are often a lot more kids than grownups, and in many pictures the grown-ups are completely absent. Is that the message people really want to send? It’s pretty rare that one sees the reality of white people learning from brown people, or the success of brown people in all kinds of economic (not to mention artistic and social) endeavors without the help of whites. Rurality and remoteness are often emphasized far beyond the reality of rapidly urbanizing and technifying countries.


I doubt that those interested in global health who post these images really mean to say that people in Haiti are orphans or that people in Togo are incapable of helping themselves. These are just visual mistakes. But it’s time to get better with visual representation. Competence in representation starts with the representation of competence.

Screenshot 2016-04-22 11.20.30

Leave a Reply

Your email address will not be published. Required fields are marked *