Humanity is not data-driven

There is much concern about the “Learning Gap.” The learning gap is really two undeniable things: a shame for our purportedly egalitarian society and a measurable fact. The ‘shame’ is in many aspects comforting and the ‘fact’ bears all the weight of the fact that there are three sheets of paper sticking out from under my computer monitor. We choose shame; it’s a feature of our Judeo-Christian cultural origins. We worship facts because they are sure and fixed and immutable. And the Learning Gap is characterized by the gap between the objectivity of the facts—data, and the subjectivity of the shame it engenders.

The ninety-second of the Roman Catholic dogma, one of a set of beliefs treated as fact for nearly two thousand years, states, “Original sin is transmitted by natural generation.(1)” We have been working on our inherent shame, treated as fact, since the conceptions of Cain and Abel. Meaning no disrespect for the great good done by Catholicism toward alleviating pain and suffering, The Church has like so many power structures simply cultivated, if not having actually manufactured, a need that it was prepared to meet for a price. Our hereditary shame, our original sin, deprives us of eternal bliss, but the Church provides an avenue to redemption. We need only do a few simple things: admit our undeserving state, accept the course our spiritual leaders offer and behave as we are told. And it’s not just Catholics. Is it?

We must admit to our sin in creating the Learning Gap. And they tasted of the fruits of class privilege and they knew their sin. We must accept the sanctity of the education reform movement. And on the seventh day they will be tested. And we will be redeemed. And the winged graduates ascended into college. And the “big data”(2) pushers should like this neat pattern correlation too. They ‘discovered’ the Learning Gap correlation, after all.

I say “discovered” because I don’t want to go right to the heart of the problem yet. You see, data are facts, and like the number of sheets of paper on my computer desk, they ‘mean’ nothing. When data are gathered, they can be sorted and arranged to create patterns, which in ‘data-ese’ are called correlations. Just before B goes up, A goes down—every time. Correlative fact, no cause, no opinion, just fact. This is the language of statistics, and we remember what was said about statistics, “Select the data that tells us what we want to know.” Could this be the case with the Learning Gap correlation? Could it be that the reason we find a difference in the performance in one racially defined group students from another racially defined group students is because something is or was going on to cause the difference? No. Because there is no cause to correlations, only data patterns. But we want things to have reasons.

“Why,” we ask, “are we here?” not just “Are we here?” Even Church dogma starts out by using the fact that we can ask as proof that there is a reason. Here’s what David Books says, with which I concur, in the New York Times, 16 April 2013:

“…I’m trying to appreciate the big data revolution, but also probe its limits. One limit is that correlations are actually not all that clear. A zillion things can correlate with each other, depending on how you structure the data and what you compare. To discern meaningful correlations from meaningless ones, you often have to rely on some causal hypothesis about what is leading to what. You wind up back in the land of human theorizing.”

Brooks contends that we seek meaning even in the meaningless and cause for the effect, and we do. It is probably deeply rooted in our psyche.

So why is the Learning Gap occurring between white students and students of color? Well, it is: no question about that. But that’s the learning gap: small “l,” small “g.” Why isn’t there isn’t the Learning Gap (capitalized) among rural, suburban and urban students, or between rich and poor students, or among the states or anywhere else that there is a gap? Well, we use race because that gap sticks to our inherent shame, our unresolved racial discrimination that we so proudly – no wait, make that, shamefully – celebrate in this country year after year. So shame can be made useful.

And the policy makers of today, like the Church leaders of the past, know how to leverage their influence and shepherd the sheep. “Close the Gap to relieve your shame.” But it may not be in the interest of that leadership to resolve the Learning Gap; it may be more useful to keep it in play. How much top down management, often in the form of cost containment, has sprung from the Learning Gap Card? Where does the power rest in dealing with the Learning Gap? And here’s the big one – If the Learning Gap were actually closed, what meaningful goal will have been achieved? How will it be more that a statistical non-correlation of data? Will the color lines go away? Will wealth be distributed more equitably? Will opportunity be truly equal? Will the nation become de-Balkanized?

_____

Those who know me know I have been railing against the obsessive elevation of data, the passionate collection to these completely dispassionate pieces of stuff, often with no prior purpose, and the religious commitment to the value of any correlation divined in the data for years now. I am increasingly convinced that data, as the raw material of the Information Age, can be capitalized, that it can be used to our benefit, or abused to cost. Moreover, similarly to iron or coal or oil, data can be manipulated to greedy ends in this Age of Greed.

Don’t let anyone tell you that the data are indisputable and non-judgmental. The choice of which data to bring forward is certainly disputable and the judgment about which correlations to divine must be highly suspect for hidden causes. Humanity is not data-driven.

_________________________________________

(1)   Loughnan, F. John. Dogmas of the Catholic Church, The Divine Work of Creation, The Doctrine of Revelation Regarding Man or "Christian Anthropology," Revised Feb. 16, 2001. retrieved from http://jloughnan.tripod.com/dogma.htm, 16 April 2013.

(2)   Mayer-Schonberger, Viktor and Kenneth Cukier. Big Data: A Revolution That Will Transform How We Live, Work, and Think, Eamon Dolan/Houghton Mifflin Harcourt; 1 edition (March 5, 2013).

Advertisements