When we use language to change the focus

From Race gaps in SAT scores highlight inequality and hinder upward mobility by Richard V. Reeves and Dimitrios Halikias.  The report is from 2017.  Nothing new in the report which hasn't been found dozens of time here in the US and internationally over several decades.  

I only came across the report because I was checking a citation in another analysis.  But having clicked through, and scanned the text, I came across this intriguing statement.  They are discussing the material and persistent SAT gaps recorded between different ethnic groups in the US.  

These gaps have a significant impact on life chances, and therefore on the transmission of inequality across generations. As the economist Bhashkar Mazumder has documented, adolescent cognitive outcomes (in this case, measured by the AFQT) statistically account for most of the race gap in intergenerational social mobility.

I am not disputing their data.  I am intrigued by the linguistic formulation.  There is great sensitivity of course around this field owing to people often being unable to distinguish between the individual and the average for any arbitrarily chosen category of identity.  There are genius and dullards at the end of every spectrum in every identity group.  Only the averages and the distributions differ and the averages and distributions say nothing about the individual.

Look at that first statement.  "These gaps have a significant impact on life chances."  That is substantially not true.  The gaps are a product of the measurement of an underlying set of capabilities.  That measurement of capabilities is useful because, for the individual, they are usefully predictive of future capability in several domains.  Not all domains and not a perfect correlation.  But usefully true.

It is the individual differences in capabilities which have a significant impact on life chances.  Not the gaps themselves between groups.  

Life outcomes are substantially a product of capability (physical and mental), cultural (what are appropriate and useful life goals), and behavioral (what are useful and appropriate means of achieving those goals) within a particular context and set of circumstances.  

Any student in the US scoring an 800 on the math SAT is going to have dramatically better opportunities than everyone else.  And the difference in life outcomes for all those scoring 800, regardless of race, are not likely great and almost certainly entirely accounted for when you control for Culture and Behavior.

The researchers almost certainly know all this.  So why do they claim that it is the existence of gaps between groups and outcomes when they know that it is the difference in capabilities which explain differences in outcomes.

I suspect this is primarily a combination of sensitivity and almost willful fuzzy thinking.  If we obscure the distinction between gaps in individual performance based on differences in individual capabilities and the reality of differences (gaps) in group averages (regardless of a category type; it doesn't have to be race),  then we create the illusion of agency and policy outcome.

In other words, if we acknowledge the role of capabilities, cultures, and behaviors in the achievement of individual life outcomes, then we acknowledge that eliminating differences in life outcomes requires the elimination of differences in capabilities, cultures and behaviors.  An almost impossible agenda.

But if we can pretend that the differences in life outcomes are due to something other than differences in capabilities, cultures and behaviors, then we have an entirely different set of possible solutions.  Maybe the tests aren't accurate in terms of measuring capability.  Maybe they aren't as predictive as we think.  Maybe they are susceptible to different social policies.

By focusing on abstract concepts like group test performance gaps and inequality, it allows us to focus on social policies which are more fun to implement than were we to try and change inherent capabilities, cultures and behaviors.  

Which is fine if our goal is to make the feelings and self-esteem of researchers and policy mavens the measure of success.

But if we are concerned with making the individual lives of individual people better, then all we are doing is avoiding that which could make a positive difference.