Standard Deviations

Gary Smith, “Standard Deviations: Flawed Assumptions, Tortured Data, and Other Ways to Lie With Statistics,”  Overlook Duckworth, 2014, 297 pp.

Author Gary Smith is a professor of economics at Pomona College and well qualified to write this book.   Although through my past reading, I am familiar with almost all the examples he uses regarding abuses of statistics, this is a good book for anyone not already familiar with the subject matter.  One tidbit this book points out that I had not considered before, is the motivation behind publishing nonsense.  It appears that much of it comes out of Universities because of their publish-or-perish policies.  I have long believed that compensation drives behavior and here is another example confirming this.

Some interesting items:

*Did you know that baseball players whose names begin with the letter “D” are more likely to die young?

*Asian Americans are most susceptible to heart attacks on the fourth day of the month?

*Drinking a full pot of coffee every morning will add years to your life, but one cup a day increases the risk of pancreatic cancer?

*there is no scientific evidence that living near or under a power line is any danger; and no-one can forestall death until after a birthday or holiday.

All of these “facts” above have been argued by credentialed researchers and backed up with reams of data and convincing statistics. Author Smith walks us through the various tricks and traps that people use to back up their own crackpot theories. Today, data is so plentiful that researchers spend precious little time distinguishing between good, meaningful indicators and total rubbish.

Among the reasons we're misled by statistics, Smith notes, is because we're hard-wired to look for patterns. "Going back to our ancient ancestors, there was survival value in recognizing patterns. The rustle in the bush might be a predator. Dark clouds bring rain. Big data and powerful computers feed into that. Because it's so easy to find patterns and we're so drawn to the patterns.  The starting point has to be common sense.

My Notes:

P. 13:  Natural selection and evolution is the reason why many people are led astray by patterns.  Our ingrained preference for symmetry is an example of how recognizing patterns helped our human ancestors survive and reproduce in an unforgiving world.  Dark clouds often bring rain.  A sound in the brush may be a predator.  Those distant ancestors who recognized patterns that helped them survive passed this aptitude on to future generations.  Through countless generations of natural selection, we have become hardwired to look for patterns and to think of explanations for the patterns we find.  Unfortunately, the pattern-recognition skills we inherited are ill-suited for our modern lives, where the data we encounter are complex and not easily interpreted.  We believe stories simply because they are consistent with the patterns we observe and, once we have a story, we are reluctant to let it go.

Pg. 14:  We are too quick to assume that meaningless patterns are meaningful when they are presented as evidence of the consequences of a government policy, the power of a marketing plan, the success of an investment strategy, or the benefits of a food supplement.  Our vulnerability comes from a deep desire to make sense of the world, and it’s notoriously hard to shake off.

Pg. 35:  Backward-looking studies often (probably always) have survivor bias in that, when we choose a sample today and look backward, we only see the survivors.  Medical histories of the elderly exclude those who did not live long enough to become elderly.  Corporate financial histories exclude companies that went bankrupt.

Pg. 66:  Author Smith refutes the claim by Donahue and Levitt that Roe v. Wade and resulting increase in abortions lowered the crime rate by as much as 50 percent.  The data shows the murder rate for 14-17 years old beginning in 1987 when those born in 1973 would turn 14 and who would turn 18 in late 1991, continued to increase during these years and did not peak until 1993.  The murder rate was three times higher in 1993 than in 1984, even though abortion was legal and readily available when the 1993 cohort was born, and illegal and difficult when the 1984 cohort was born.  The real story is that the murder rate peaked for all age groups in the early 1990s.  Whatever the explanation, legalized abortion was not the cause.

Pg. 86:  In 1981, interest rates reached 18 percent on home mortgages. The unemployment rate rose from 5.8 percent I 1979 to above 10 percent in 1982, the highest level since the Great Depression, but the Fed’s policy reduced the rate of inflation from above 13 percent in 1979 to below 4 percent in 1982.  Prices increased at an annual rate of 10.7 percent during Carter years and 3.7 percent during the Reagan years.

Pg. 152:  The DOW often substitutes stocks when it is judged that listed stocks are less representative than another in representing the market as a whole.  Yet, those stocks deleted from the DOW 30 often do better in the ensuing years than the stock it replaced:  A 2006 study of all 50 changes in the Dow back to October 1, 1928, when the Dow 30 stock average began, found that deleted stocks did better than the stocks that replaced them in 32 cases and did worse in only 18 cases.  A portfolio of deleted stocks beat a portfolio of added stocks by about 4 percent a year.  This is an excellent case of regression to the mean.