The Data Says So … Doesn’t it?

The Data Says So … Doesn’t it?

The paradox of alternate realities

Be careful out there!  It’s been a very long time since I slogged through my social statistics classes as an undergraduate Sociology major.  But, one thing I learned that stuck with me is a simple truth.  If you’re good with the math you can make the data say whatever you want it to.

Case in point…

The ongoing, regular and constant conversations about disparity in gender and race amongst the ranks of elite classical musicians in American orchestras.  When you look at the data you see a world dominated by white Anglo Saxon men.  Sure, you can debate and argue nuances involving the rise in the numbers of Asian men and women, or whether or not you should make distinctions between western and eastern European origins.  But why bother?

The fact is – most American orchestras have been practicing blind auditions for decades.  This means that auditions are held behind a curtain.  The path to the performance spot on the stage is carpeted to avoid the sound of heals on wood.  The applicant is not allowed to speak to avoid exposing their gender through the sound of their voice.  The applicants audition is judged on artistic merit only.

Yet, the numbers aren’t wrong.  Facts are facts.  So, the pundits go on and on about the glass ceilings that cause the social inequity that is enumerated by the data.  But, they make a very common mistake.  They confuse correlation with causation.  The data certainly correlates.  But that doesn’t mean that one thing causes another.

For example – the number of African American women is disproportionately low compared to the general population, and the magnitude of the difference is statistically significant.  There is a data correlation.  Many may assume that there is a social barrier that exists that must be influencing the balance which results in the data we see.  But, it’s equally possible that the real culprit is not a negative limiting factor, where one group is actively excluding another.  It’s possible that the root cause is sociological and a result of a simple fact.  The underrepresented group has less desire to be counted among the group being measured.  They exclude themselves because they have no interest in being included.

Of course, we can debate why this is so, and I’m sure someone will propose that the reason they don’t aspire to be among elite classical musicians is because some other group in society oppressed them systemically which resulted in a shifting of cultural values that plays out as a lack of interest in the classics.

As you see, we can paint the picture any way we like using the existing data if we’re clever and committed to proving our position.

Outside the realm of data analysis and statistics we often see that this same phenomenon holds true.  Reality is dependent on the perceptions of the observer.  Multiple people can behold the same object or event yet ascribe completely different meaning to it.  Much like elusive quantum particles that can only be detected when observed – the act of observing disturbs and influences the particle – reality can be disturbed by observation.

Multiple possibilities.  Multiple observers.  Multiple realities.  All in conflict with one another, yet all correct.