Subscribe to DSC Newsletter

Above - During a Session at the Archives of Ontario in 2011

In this blog post, I describe my early experiences leading me to conclude, data as we know it tends to be "disembodied" - that is to say, often lacking any kind of connection to different types of bodies.  When we talk about things being disembodied, I suspect some form of decapitation is involved.  However, I am referring to the disembodiment of information.  There are some consequences to disembodiment - such as causal disassociation - that can reduce the usefulness of data-mining.

The image above shows my surroundings while reviewing the contents of about a dozen boxes containing management records.  I was accessing data from an organization that no longer exists although it provided service for about 30 years.  It operated before the advent of desktop computers; and it was closed down soon after computers became common in offices.  A contemporary human resources department operates with a lot of more data these days; so some might argue that my observations are a bit dated.  However, perhaps current conditions can be best explained by examining the early contributing factors.

The organization dealt with a lot of workers.  When it was first conceived in the 1960s, there was great interest on employees as individuals - specifically those that had drinking problems.  The organization later became responsible for all sorts of counselling needs.  In the boxes, I found photo-albums.  When a person has quite a lot of data, it is sometimes easy to lose sight of the underlying reality behind the data - namely, the people.  I consider it important to be aware of the reified nature of the portrayal.  I believe "reification" is a term that has been attributed to Karl Marx.  I think reification is supposed to give rise to a type of alienation.  I found it difficult while staring at these people with hair and clothes from the 70s and 80s to disassociate their bodies from their metrics.

A profile of insurance claims that I found in one of the boxes suggests that about half of the claims related to muscloskeletal and cardiovascular problems.  I mention these two claim categories specifically because the proposed cost-cutting solution was unlikely to lead to a reduction in claims:  decentralization.  Not better chairs and desks.  Not adjustable monitor stands.  Not more opportunities to step outside and get air.  Decentralization, a non-health approach, was regarded as a way to tackle rising disability claims.  I am not saying that decentralization is necessarily an inappropriate path in terms of administrative developments for an organization; but it might be unrelated to the underlying issues.  If a person had an addiction to alcohol, and somebody suggested that decentralization could help, I would certainly express some concerns about the reasoning.

I have noted therefore that an abundance of data does not necessarily equate to a comparable amount of useful guidance.  Today we have an increasingly abundant supply of data.  The use of data-mining techniques can potentially create a new age of data for analysts and researchers.  Collecting this data represents the beginning of a process rather than its end since it is disembodied data - decapitated early in life and cursed to wander aimlessly in the digital labyrinths of hard-drives.  Neither reality nor interpretation resides in data.  In my example, data provided a bridge for decision-makers to come to conclusions that were disassociated from the lived truths behind the numbers.  Can any instrument of conveyance provide anything more than an assertion of reality?

I actually support the data-mining movement.  Apart from the meaning of the data under examination, I am actually interested just in data itself as a subject.  Nonetheless, there is social obligation.  A person has to be committed to truth and beneficial outcome.  The issue of poor or faulty embodiment should be important since actual bodies are involved.  These bodies might be natural habitats, municipal infrastructure, students, and health-care recipients.  Lack of embodiment in data can result in inappropriate conclusions, uninspired ideas, and unproductive approaches.

Any and all comments are warmly invited.

Views: 203

Tags: analytics, critera, data-mining, devolution, disembodiment, efficiencies, health-care, metrics, neoliberalism, organizational, More…prescriptive, reductive, strategies, systems, workplace

Comment

You need to be a member of Data Science Central to add comments!

Join Data Science Central

Videos

  • Add Videos
  • View All

© 2019   Data Science Central ®   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service