Saturday 9 September 2017

Freakonomics - Part 2

FREAKONOMICS - Part 2

by Steven D. Levitt & Stephen J. Dubner


Chapter 4: "Where Have All The Criminals Gone?" of Freakonomics provides strong support for the 3rd concept that co-authors Steven D. Levitt and Stephen J. Dubner explore in their best-selling book.  Concept 3 is Dramatic effects often have distant, even subtle causes.  Chapter 4 also reaffirms their 2nd concept - Conventional wisdom is often wrong - and their 5th concept - Knowing what to measure and how to measure it makes a complicated world much less so.

In Chapter 4, the authors set out to show that several commonly held views for why the crime rate dropped dramatically during the 1990s don't hold up to scrutiny.  For instance, the most popular reason given for the crime drop (as suggested by the LexisNexis database*) is innovative policing strategies, such as those instituted in New Youk City by Mayor Rudolpho Giuliani and his hand-picked police commissioner William Bratton.  The authors conclude though that "...a careful analysis of the facts shows that the innovative policing strategies probably had little effect on this huge decline." (128)   For one thing, the New York City crime rate was alreading plunging even before Giuliani and Bratton's arrivals on the scene.  Furthermore, as it turns out, crime rates dropped significantly across all of America in the 1990s, not just in places such as New York City that introduced innovative policing practices.  

Levitt and Dubner also present other commonly given reasons to show that other conventional explanations for the crime drop (ie. tougher gun laws, a stronger ecomony) likewise don't adequately explain why the crime rate dropped during the 1990s.

Their analysis reveals instead a hidden and seemingly remotely connected reason: the legalization across America in 1973 of abortion:

            In the early 1990s, just as the first cohort of children born after Roe v. Wade 
            was hitting its late teen years - the years during which young men enter their
            criminal prime - the rate of crime began to fall.  What this cohort was missing,
            of course, were the children who stood the greatest chance of becoming
            criminals. (139)

In summary of Concept 3, the authors write:

            We have evolved with a tendency to link causality to things we can touch or
            feel, not to some distant or difficult phenomenon.  We believe especially in
            near-term causes. (140)

In answering, in Chapter 5, the question How much do parents really matter when it comes to success for their children in school?, the authors again provide support for their 2nd and 5th concepts. To address this question, they apply regression analysis (which determines correlation) to data obtained from the U.S. Department of Education's Early Childhood Longitudinal Study.  The results are surprising.  Conventional expanations such as The child's family is intact, The child attended Head Start, and The child's parents read to him [sic] nearly every day do not correlate with success for a child in school.  Furthermore, "Despite the conventional wisdom, watching television apparently does not turn a child's brain to mush...", that is, it doesn't correlate with a lack of school success.   According to the survey, the factors that do correlate with early school success are characteristcs of the parents, such as their being well educated, successful in careers, and healthy.

What the authors conclude from they analysis is that "...it isn't so much a matter of what you do as a parent [that accounts for a child's success in school]; it's who you are. (178)  Add Levitt and Dubner:

            For parents - and parenting experts - who are obsessed with child-rearing 
            techniques, this may be sobering news.  The reality is that technique looks
            to be highly overrated. (177-178)


In line with their 5th concept, the authors analyze the "predictive powers" associated with first names that parents select for their children.  They state that "Many parents seem to believe that a child cannot prosper unless it is hitched to the right name." (181)   Thus, they ask the question Does the name you give your child affect his life?  

The California birth-certificate data for 1961 (more than 16,000,000 births) indicates that, overall, people with distinctively black names (ie. DeShawn or Imani) do have a worse educational, income, and health outcome than people with white-sounding names (ie. Jake or Molly).  However, according to the authors, a deeper analysis of the data reveals that this correlation doesn't equal a cause and effect relationship.  Rather, they conclude that a black or white sounding name is only "...an indicator - not a cause - of [an individual's] outcome." (192)  They elaborate as follows:

            If two black boys, Jake Williams and DeShawn Williams, are born in the same
            neighbourhood and into the same familial and economic circumstances, they
            would likely have similar life outcomes.  But the kind of parents who name their 
            son Jake don't  tend to live in the same neighbourhoods or share economic
            circumstances with the kinds of parents who name their son DeShawn.  ...A
            DeShawn is more likely to have been handicapped by a low-income, low-
            education, single-parent background. [the actual cause] (191)     

The authors' answer then to the question they pose is clear:

            What the California names data suggest is that an overwhelming number of
            parents use a name to signal their own expectations of how successful their
            children will be.  The name isn't likely to make a shard of difference. (207)


      The LexisNexis database cites reasons offered in articles found in America's 10 largest 
         newspapers.