​

“It’sa dangerous business, Frodo, going out your door. You step onto the road, and if you don’t keep your feet, there’s no knowing where you might be swept off to.”

― J.R.R. Tolkien, The Lord of the Rings
It is likewise a dangerous, or perilous might be a better word, business to begin to count things. Once you start counting, you never know where it will take you. What you do know, or at least what you should know, is that where it takes you is highly influenced by your assumptions in place when you began to count.

Counting is fundamentally an act of control. We seek to understand and thereby control our environment or our relationship to our environment. We seek to control things by counting in order to manage their exchange or monetizing the things being counted. We control people through counting in a number of ways, not the least of which is to define groups in and out of existence.

The notion of “information justice” captures this latter idea. Jeffrey Alan Johnson defines it thusly in an upcoming text:

“Information justice refers to the fundamental ethical judgment of social arrangements for the distribution of information and it’s effects on self-determination and human development. It is a broader subset of the notion of political justice, applied to questions of information and information technology.”
And so my challenge of how I think about and how i perform counting and analytics is this.

How do we count without doing harm, without taking away voice and recognition from minority groups, and how do we use analysis to improve outcomes without stealing it denying agency of those we claim to serve?

So, I have a few principles that I follow.

1) Counting must give safe voice to the voiceless. We can’t allow counting to become a tool of targeting subpopulations. We must think about the act and processes of counting to protect confidentiality as appropriate.

2) Counting and reporting should illuminate and guide decision-making in non-harmful ways. By this I mean that data and information displays should follow good practices of clarity and appropriateness of context and completeness. There should be clear statements of the limits of any measure. Average wages of graduates by itself has so much potential used by itself to be completely unrepresentative of 80% of people reported under such a measure. Distributions matter.

3) Analytic models that result in risk factors that include race/ethnicity or gender are indicators that you are measuring against the students or clients you wish you had, not those you have. We should delve more deeply into understanding the relationship between what we do and who we think we are serving.

4) Slick displays and data visualizations are too often like air brush paintings where the artist can’t draw. Slickness is often used to hide the lack of substance. Simplicity and basic displays are better for highlighting difference or the lack of distance.

5) There is often no other more powerful act than the act of definition. In the absence of definition, creating and publishing one is powerful, even without fanfare. A definition may seemingly lay dormant on a web page for years and then appear in law. It’s powerful and a tremendous responsibility.

I guess to sum up I would say this.

I count to give voice and support to those we serve, striving to do no harm, and to never restrict nor deny agency to those being counted. The analytics I perform are to improve systems’ and institutions’ understanding of self and to support individuals in finding the excellence in the lives they desire, within the domain of my work.