You are currently browsing all posts tagged with 'generalizability'.
Displaying 1 entry.

Generalizations

  • Posted on September 17, 2014 at 10:00 AM

So, a basic assumption of statistics is that you can create a framework, select a random sample, and produce survey results that are generalizable to the general population. This has been asserted as a fact in every attempt I have made to study statistics. This time around I’m admitting right off the bat that I don’t buy it.

Maybe it’s that I don’t understand where this “fact” came from and how they reached the conclusion that it’s true. More likely, it’s the “fact” that I am and always have been something of an “outlier.” You cannot talk to someone who has the same age as me, the same gender as me, the same race as me, and who is living in the same area as me and conclude that their views (the answers to the survey) can be generalized to me.

This assumption of generalizability leaves out far too much that is of personal significance:

  • Would she have children?
  • Would her children have disabilities?
  • Would those disabilities include autism, sensory processing disorder, and epilepsy?
  • Would her children be teenagers?
  • Would she have a disability?
  • Would that disability be fibromyalgia?
  • Would she own her own business?
  • Would she be going to graduate school?
  • Would she already have a graduate degree?
  • Would she have dabbled in politics long enough to become disgusted with our system?
  • Would she vote despite that disgust?
  • Would she make every effort to be an informed voter?
  • Would she have experienced life below the poverty line?
  • Would she have struggled to lift her family out of poverty?
  • Would she have experienced life as a married teen mother?

I could go on and on. The point, however, has been made. The things that shape my answers cannot be attributed to my race, my age, or my gender—at least, not exclusively. My experiences—who I am—shapes how I see the world. And that shapes my answers to surveys. While I may be an outlier, I’m pretty sure the same is true of everyone else and that those generalized factors aren’t the major determinants of their experiences.

How can we possibly generalize people based on random attributes when those attributes aren’t really what makes them who they are?

Part of me wonders if inequality based on age, race, and gender persist with such prevalence because our society, at its core, still believes that those attributes really do determine who and what we are.