Saturday, October 14, 2006

In Defense of Data: Long, Meaty, and Raw

Background: I have a major in Economics, a research job, and an interest in Sociology, Demography. I spent about twenty hours this week doing data programming. Also, if this were a working paper in my field it would read in big block letters:
PRELIMINARY AND INCOMPLETE: COMMENTS WELCOME.

A while ago EL of My Amusement Park posted about the New York City's board of health allowing transfolk to get new sex-corrected birth certificates, and asked an important question about why we feel to need to ask that question there. I posted a response. And so did Jenn. You can take a look at the whole exchange here. Though I will also quote below.

Jenn wrote:
I know demographics are supposed to be terribly helpful. Like, they're supposed to help the medical establishment deliver better health care... to those tragic white men who can't get their peckers up. I know demographics help the government figure out what "minority" groups really need, so they can withhold it until they get enough vote service. It helps teachers know which kids to attend to and which are destined to fall through the cracks anyway and aren't worth wasting time on. And don't forget how it helps businesses determine that yes, Virginia, only the needs of white male consumers matter.
Without demographics, we'd have to go back to plain old KKK style bigotry. It's so much nicer to be able to couch it all in lots of scientific-sounding numbers (that we've manipulated to support our pre-existing POV).
Bigotry is dead! Long live Bigotry!
What does knowing someone's gender really tell the govt that it needs to know? I can't think of anything except whether the person deserves first-class treatment or second.
And I understand her point. Demographics can be **terribly** helpful. They are a lynchpin in the bio-political power that the modern nation state has had in the last one hundred and fifty years. Now, some theory:
Bio-power is one of the many technologies of power that was elucidated by Michel Foucault. In History of Sexuality, Vol. I. he describes “the emergence in the field of political practices and economic observation, of the problems of birthrate, longevity, public health, housing and migration. Hence there was an explosion of numerous and diverse techniques for achieving the subjugation of bodies and the control of populations” (Foucault, 140). It is closely linked to bio-politics, which David Hoy defines as “the strategies that are to be pursued in implementing bio-power”; these two are so linked that at points one of them will be used to signify both (Hoy, 74). This “management of life” permeates into people’s private lives through the techniques of bio-politics that govern people’s sexuality and other aspects of their biological existence (147). Sexuality is more intensely regularized in this framework because it is at “the juncture of the ‘body’ and the ‘population’” which are the main targets of disciplinary and regulatory power respectively (HOS 147). He describes the place of sexuality in a bio-political framework thus, “Through the themes of health, progeny, race, the future of the species, the vitality of the social body, power spoke of sexuality and to sexuality; the latter was not a mark or a symbol it was an object and a target. Moreover, its importance was due less to its rarity or its precariousness than to its insistence, its insidious presence” (147-8).

The fact that bio-politics creates a poor population that is inherently “lazy, indigent, and undeserving” blinds those with who have privilege from interrogating the structures of poverty in their society. As Foucault notes the divisions in society take on a biological rather than political level. This allows those with privilege to be politically complacent and ignorant. They are permitted to think about poverty as the problem of a population rendered as other through appeal to unchangeable characteristics. Because this is the dominant view, mobilization does not occur in resistance to the techniques of bio-power. This complacency breeds more privilege and the perpetuity of those ‘unchangeable’ characteristics. This social blindness and abdication of responsibility is one of the most pernicious effects of the bio-politics that surround welfare policy and poverty in the United States.
And I get all of that, heck I wrote that above "block-quote-bigger-than-something-big-in-your-vicinity". But I also think that data has some advantages that I break into two main sections (for now).

1. The Master's Tools will Never Dismantle the Master's House
And yet, they are what we have. They are like props for some grand improv theater game, and while I think that other tools will be created, imagined and will become powerful, I also am not quite ready to throw the baby out with the bathwater. I also think that the "Master" probably doesn't exist anymore: what we have is whole cadres of people who are trained to think like him who may or may not agree with him really -- but only know how to talk about things that the way that they were taught. I want to talk to them. A lot of scholarship in Sociology and Feminist Economics is using these tools, and using them to ask the questions in different ways. [examples will be included in full version~]

2. Narrative is Also a Trap: Pixillating the Narrative Trope
The only opposition in the world is not: Data v. Narrative. But let's pretend for a moment that it is:

The discourse of confession and truth-telling has made the creation of a personal, individual narrative an imperative. Often this personal narrative follows a pattern that is edging on trope whether it is a story of birth, death, or coming of age. Because of this patterning, the narrative may not be liberating or reflective of much, as was noted by Eve Sedgewick in The Epistemology of the Closet. In this way the individual, in a true sense, is lost in what the current age rabidly declares is an expression of that very individualism. Part of this mythic belief in the personal narrative is the idea that any method of truth gathering that denies the individual the opportunity to tell their own story is inherently oppressive, veiling, and authoritarian. These methods, typified by data collection, reduce the ‘individual’ to a mere number, masking the nuances of their experiences.

In this scenario the individual was lost to the collective, and individual lives were created and ended on the basis of the desires of a racist state with police powers. This power dynamic can have significant pernicious effects on populations, as evidenced by institutions like public assistance programs and AIDS funding policies in the United States.

Generally the very methods of data collection feed into the power of the state, limiting the ways in which people can identify and describe themselves and their families. I faced this in a very tangible way when I realized that the data set that I was using for my thesis did not allow respondents to label themselves as either more than one race or as Latino/Hispanic (decidedly, and happily, NOT the case with census data). Another example is the way in which sex and gender identification are constantly regulated through data collection that only allows people to identify as “male” or “female”. These racist and hetero-normative assumptions about the ways in which respondents will identify and respond clearly truncate the potential for liberating truth gathering. However there is no requirement that questionnaires be constructed in this manner, they could allow for a much broader range of responses with only minimal inconvenience to the statisticians who analyze the data. While there is major institutional foot-dragging to prevent this from happening, it is possible. Ultimately more accurate and precise measurements are always in the interest of the researcher because it means that they can state with more confidence that their findings actually mean something.

If we return to the problem of the narrative as oppressive because of the way in which it compels subjects to follow a preset trajectory, what are the possibilities associated with looking toward data collection? Data collection has many problems, yet it is very effective at breaking up the stories that people tell about their lives. Instead of asking people if they are healthy, there is a series of questions that ask about specific behaviors that a healthy person can do in a day, such as climbing a flight of stairs, and then codes them along a preset scale. While these scales, particularly of health and fitness, may be very subjective and problematic there is a value to the way in which people’s experiences are pixilated – broken, disintegrated, and potentially distilled. The conscientious researcher could rearrange these pixels, creating them into an aggregate picture that was used not to oppress or regulate, but describe the lives that people are living in ways that they may not be able to articulate. Once these new articulations are provided, the possibilities for uses for the new information abound. This is particularly true where the stories that people tell about themselves serve as barriers to the building of coalitions and communities. If there were data explaining the common problems between various populations with the state policing their family structures it is possible that a coalition would be more forthcoming. Whereas without that data, strong narratives of gender, sexuality, race, and class might prevent those coalitions from being built.

Further, often the narratives are not personal, but rather are cultural. I think probably the best example that I can give of this (without revealing where and for whom I work, and what I'm working on) is the narrative about the middle class. The middle class in America is constantly thriving - everything is constantly getting better. This is the story that is told and that we tell ourselves and often believe. However, in the last few years this has broken down -- crumbled even - why? Data. Data that says that homeownership is a trap that is just as likely to put you in debtor's prison as deliver you to a white picket fence. Data that says that the rich are, in fact, getting richer, and the poor are getting poorer, and everyone else is just scrapping by. And folks are starting to not believe quite as much anymore Do you need to know people's gender to know that this is happening? No, you don't, but you do need it to talk about the ways that elderly women are more likely to be poor than elderly men because pension benefits screw them over (if you have access to JSTOR or other academic loveliness - do yourself a tiny lil' lit search, if not take my privileged word for it).

Ok. I'm done. I'm sitting in a coffeeshop in LiberalCity, RedState, with TP. And I have some work that I brought "home" with me over the weekend, not to mention a novel to read, and a person to kiss. Congratulations for making it this far.
___________________________________
Dork-out further here:

*Foucault, Michel. The History of Sexuality: An Introduction. Translated by Robert Hurley. Random House, Inc., New York, 1978.
*Foucault, Michel. Society Must be Defended: Lectures at the College de France 1975- 1976.Translated by David Macey. Ed. Mauro Bertani and Alessandro Fontana. Picador, New York, 2003.
*Hoy, David. Critical Resistance: From Poststructuralism to Post- Critique. MIT Press, Cambridge, MA.

~Said version may never be written.

2 comments:

ben said...

i simply don't know how we'd organize ourselves if we didn't have the ability to do social research in this fashion...

the problem, like in other fields, is who is doing the research, what kinds of questions are they asking, and do they fully represent the complexity of the subject matter?

the furor over the iraq body counts has taken a chance to actually think about the sheer scope of devastation we inflicted in to a "math is too hard, so we couldn't have done it" anti-intellectualism.

so three cheers for socially conscious statisticians!

greymatters said...

I just wanted to pipe in to say I did read this, and am pondering. I'm not sure what to say, largely because I'm someone who does that stats thing (more cliometrics than econometrics), and who does not eschew narrative.

Interesting, C. Interesting.