28 November 2010

Gallimaufries...

I have an anthropological friend who has a fine line in tart put downs when it comes to survey-based quantitative social research. "Isn't it mad?" she asked me recently. "You take a series of stupid questions, ask them hundreds of times, shunt the answers through statistical analysis - as if sheer multiplication could make them fundamentally any less stupid." As I say, she's an anthropologist, so her rich descriptive prejudices may be forgiven and her venomous ethnographic saws are always worth gleefully and flippantly reviving when one meets a soor-faced economist who counts numbers amongst her closest cronies. One for those of you who enjoy a percussive thwap as a descending lead balloon ricochets off the floor.  That said, her doubts are substantive and raise important questions for anyone conducting or reading a quantitative analysis.  What questions have you asked? What answers do you permit? Are these the right questions, reflective answers? I'm reminded of one of my lecturers in my days at the University of Edinburgh, who asked various undergraduate law students How would you assess your intelligence on a scale of 1 to 10? The woman who instantly said "ten" certainly afforded an insight into her psyche. Most said "seven", thinking it more-or-less struck the balance between modesty and false-modesty's conceit. Some referred to demographic data, noting with an almost apologetic tone that "statistically" they must be an eight or a nine. Other, precise-minded (probably conveyancing) types applied their own eccentric calculus to the question and decided they must be a 6.7591 exactly.

My abiding sense was of the crassness of the question, the scantiness of the answers, the debatable issues about conceptions of intelligence and the innocence of the scale - which the brute enquiry simply doesn't struggle with. And look at the various logics folk used in their answers. In the first, a delusional conceit. In the second a conspicuous consciousness for how one's answer makes one appear. Thirdly, comparisons made possible because of formalistic standards of assessment and past academic performance - and the assumption that intelligence should be unproblematically identified with these. Finally, a private double-entry brightness ledger whose calculations remain eccentrically specific and decidedly opaque.

Unlike my anthropological friend, I'm more of a friend of social research by numbers, as long as we keep a proportionate view of the data such approaches can yield - and what the process unavoidably neglects, the issues it brackets, the (potentially exceedingly problematic) assumptions it makes. A wee example can be furnished by "Weber's Sandwich" who are conducting both Scottish blog reader and blog writer surveys.  I suspect others found some of the questions rather too specific - but being helpful minded respondents, you approximate, fudge, guess, gloss over definitional problems. We've seen similiar issues raised this week on the general idea of quantitative research into happiness or well-being. Similarly, I just wanted to touch on another piece of work, the Scottish Government's recently published paper on The Experience of Civil Law Problems in Scotland 1997 - 2004. The publication tries to combine existing sources of data - which are comparatively scanty - "to provide context and a baseline for the findings of the Scottish Crime and Justice Survey ( SCJS): Civil Justice Module, which will run from 2008 to 2013". I notice that Scottish legal magazine the Firm rather misread the document, suggesting that the research reveals that 76% claimed they had been treated unfairly by the police. This seems  a scandalously high percentage - scandalous, but ... er ... false. If you look into the document, you find that the 76% figure isn't the total number of respondents experienced unfair polis treatment, but the number of those who experience such treatment who sought assistance of some kind. Indeed, this reading is confirmed in the section on prevalence, which confirms that "the lowest group range in prevalence from less than 1% to 4% ... include ... unfair treatment by police". Once before, I've mentioned what is generally called legal consciousness or legal culture research which in a classically sociological way, examines how folk conceive of law and legal institutions. Much of this is inspired by Patricia Ewick and Susan Silbey's (1998) Common Place of Law: Stories from Everyday Life which broke ground by exploring how ordinary folk from the American State of New Jersey conceived of law's place in their lives and their understandings of conflict.  The book opens with a telling description of an African American woman and her travails in court and how she only received her legal entitlements because of the intervention of her wealthy white employers. As far as I'm aware, nobody has conducted similar qualitative research in Scotland. Whatever the limitations of the quantitative view of this study and the research which follows, we ought to appreciate all attempts to understand the complexities of law's place in our communal life. Its amazing how often and in how many fields researchers refer to the extent to which  issues in Scotland's social life are under-research, whether in contemporary and historical fields. In this respect, the limited socio-legal research conducted into Scots Law simply reflects wider trends.

Finally, on an unrelated note, I've composed today's Scottish Roundup, entitled "Clinging to November", including a passing meditation on my younger self's travails on the fitba' pitch. Poor unfit mite. Read it all here.

7 comments :

  1. Try looking at the data for the Global Warming Myth, if you can find it, the raw stuff that is.

    Sleight of hand, scientifically applied and institutionalised for self enrichment.

    ReplyDelete
  2. Cabernet Sauvignon, Merlot and Shiraz up tomorrow on Newsnet.

    ReplyDelete
  3. Given the nipping and eager air which has set in across the country, it seems very seasonable to be slurping on robust reds. Anticipate your thoughts with interest.

    ReplyDelete
  4. I'll get around to Spanish reds for hand-to-hand fighting sometime in the thread.

    Are you a Port man/wumman?

    I'm off to Oporto in January and will do a piece on Port after that.

    Snowing here as well

    ReplyDelete
  5. I seem to be living in the only pocket of the country at the moment over which a single flake of snow is yet to fall. A pity really, as Oxford like Edinburgh is starkly beautiful when its geometries are blanketed.

    For the categorical avoidance of doubt, I am indeed an XY and do often enjoy a bumper of port and fierce, leathery Spanish reds. That said, at this time of year I particularly adore floral, apricot-coloured pudding wines. Delicious!

    ReplyDelete
  6. Plummy and leathery???

    OK, a Merlot from Navarra and Rioja, say 3 years old.

    ReplyDelete
  7. Come to think of it, ant red from Navarra or Rioja these days

    ReplyDelete