Correlation versus causation, in a graph

We’ve mentioned the relationship between ice cream and murders before, but here’s another correlation example.  Internet Explorer drives people to murderous rages!

47D7zGq

Source: Chris Blattman (click to visit his site)

To the often misinterpreted data of correlation, I’ll add another form of misleading evidence often found on the Internet: the personal anecdote.  I haven’t used Internet Explorer in almost a decade, and I haven’t gone on a single homicidal rampage in that time.  It must be true!

Roundup on race and IQ

Jason Richwine’s 2009 Harvard dissertation IQ and Immigration Policy has been all over the news for the past few days, after the circulation of a controversial Heritage Foundation report on immigration with which he was involved.  A lot of thoughtful people have written on the topic and provided a better analysis than I can in a short blog post. So instead, I’ve collected a reference list of a dozen links for anyone who wants to read more. I’m trying to focus on articles that discuss the science and philosophy of the issue, not politics or whether his dissertation was properly reviewed.

  • Heritage study co-author opposed letting in immigrants with low IQs from The Washington Post. This article seems to have been the spark that lit the fire of this controversy.  “Toward the end of the thesis, Richwine writes that though he believes racial differences in IQ to be real and persistent, one need not agree with that to accept his case for basing immigration on IQ. Rather than excluding what he judges to be low-IQ races, we can just test each individual’s IQ and exclude those with low scores.”
  • What We Mean When We Say ‘Race Is a Social Construct’ from The Atlantic. “When the liberal says ‘race is a social construct,’ he is not being a soft-headed dolt; he is speaking an historical truth.”
  • Why People Keep Misunderstanding the ‘Connection’ Between Race and IQ also from The Atlantic. “Among rich kids, good opportunities for developing the relevant cognitive skills are plentiful, so IQ differences are driven primarily by genetic factors. For less advantaged kids, though, test scores say more about the environmental deficits they face than they do about native ability.”
  • Two from the American Anthropological Association: RACE – Are We So Different? and the 1998 AAA Statement on “Race”  “At the end of the 20th century, we now understand that human cultural behavior is learned, conditioned into infants beginning at birth, and always subject to modification. No human is born with a built-in culture or language. Our temperaments, dispositions, and personalities, regardless of genetic propensities, are developed within sets of meanings and values that we call ‘culture.'”
  • Race And IQ. Again. on Andrew Sullivan’s The Dish. “But please don’t say truly stupid things like race has no biological element to it or that there is no data on racial differences in IQ (even though those differences are mild compared with overwhelming similarity). Denying empirical reality is not a good thing in any circumstance.”
  • Six Reasons Why Race-and-IQ Scholarship is an Intellectual Dead End from studentactivism.net, largely a response to Sullivan [via @savageminds] “Let’s say it were discovered that one American racial group was, once all the effects of nutrition, healthcare, education, income, parenting, and every other environmental factor were controlled for, on average innately slightly less intelligent than another. Would that finding justify discriminating against the less intelligent group in employment, education, or any other realm of endeavor?”
  • Should Research on Race and IQ be Banned? from the Scientific American Cross-Check blog.  “Irony Alert: It just occurred to me that two recent films, The Great Gatsby and Django Unchained, feature villains who spout pseudo-scientific theories of white superiority. The films imply that these theories are ludicrous relics of our racist past and that no modern person could possibly believe them. If only.”
  • The IQ Test from Slate. “’I have never worked on anything even remotely related to IQ, so don’t really know what to think about the relation between IQ, immigration, etc,’ Borjas told me in an email. ‘In fact, as I know I told Jason early on since I’ve long believed this, I don’t find the IQ academic work all that interesting. Economic outcomes and IQ are only weakly related, and IQ only measures one kind of ability.'”
  • Flynn, Ceci, and Turkheimer on Race and Intelligence: Opening Moves from Cato Unbound (2007). Don’t miss the links to response essays at the bottom of this article. “Take, for example, health care. Patients differ enormously in intelligence level, and these differences have life and death consequences for them. Individuals of lower health literacy, or IQ, are less likely to seek preventive care even when it is free, use curative care effectively when they get it, understand and adhere to treatment regimens, or avoid health-damaging behavior.”
  • A Talk with Jason Richwine from The (Washington) Examiner.  (speaking about making an even more controversial statement in a 2008 panel) “What I emphasized was that ethnic group differences in IQ are scientifically uncontroversial. That being said, there is a nuance that goes along with that: the extent to which IQ scores actually reflect intelligence, the fact that it reflects averages and there is a lot of overlap in any population, and that IQ scores say absolutely nothing about the causes of the differences — environmental, genetic, or some combination of those things.”
  • And finally, a 2009 Guardian review of Stephen Jay Gould’s The Mismeasure of Man. “This book should make any sensible person wary of attaching too much value to IQ tests (there’s some glorious stuff on the quixotic allotment of IQ ratings) and should make anybody very suspicious of statements about ‘group IQ’ or the presumption that some races are innately more clever than others.”

On ice cream, lead, and murder

On this blog in the past, we have looked at some intriguing ways in which social issues such as violence may be considered as epidemics.  We have also looked at some of the problems in public health with confusing correlation with causation; a classic statement of the fallacy is often given as follows: in summer ice cream sales go up, and murder rates go up.  Therefore, eating ice cream causes murder.

The example of ice cream and murder is absurd, but it points out just how difficult it can be to ascribe causation definitively in matters of public health.  Clearly, both ice cream sales and murder rates are independently affected by the same actual cause (heat waves), but one could easily imagine compelling data showing that ice cream sales go up just before each wave of violence.  And in fact, a fascinating new piece in Mother Jones has been getting a lot of attention in public health circles this week because it shows exactly that kind of compelling relationship between violence and a different factor: leaded gasoline.

Through a pretty careful analysis of past publication, the article makes an extraordinary claim: “Gasoline lead may explain as much as 90 percent of the rise and fall of violent crime over the past half century”.  But it has the data to back it up, and what’s really intriguing is that these correlations hold from the macro- all the way down to the neighborhood level.  In neighborhoods where lead is removed, crime rates drop a predictable number of years afterward.  If there really is a causal relationship between lead exposure and violent crime, we should be making the removal of lead from the environment a top priority- and maybe we should also be reconsidering the effectiveness of the police campaigns that are claiming the credit for the tremendous decline in violent crime America has been experiencing in recent decades.

But is this really ice cream and murder all over again? Scott Firestone has an excellent blog post about the MJ piece that does a nice job discussing why we might temper our enthusiasm about these findings somewhat (although he also finds the data very compelling), and it’s worth reading just to think more about how hard it can be to prove anything with certainty, even when the evidence is extraordinary (think of how successful tobacco lobbyists were for so long in creatively interpreting the data on the health effects of smoking).  There’s a brief and well-executed discussion this week in Scientific American about just how hard it can be to establish causation in health on another issue:  whether even very moderate amounts of drinking during pregnancy has any negative effects on babies.  This should be easy to establish, but it isn’t: in part because of ethical considerations (you can’t set up a control for potentially harmful behavior), in part because of the reliability of self-reports, and in part because of confounding variables like “lifestyle” associations (the same arguments tobacco lobbies make).

Cancer survivors forgoing care because of cost… perhaps

Young adult cancer survivors often forgo follow up medical care, reads the headline of this article on amednews.com (via the AMA’s Twitter feed). The article goes on to cite a study in which

Researchers analyzed data from the Centers for Disease Control and Prevention’s 2009 Behavioral Risk Factor Surveillance System on adults 20 to 39. A total of 979 had been diagnosed with cancer between ages 15 and 34 and were at least five years past the date of their diagnosis. They were compared with 67,216 adults with no cancer.

Both groups had similar rates of having health insurance. But those with a history of cancer were 67% more likely to go without care because of cost.

The rest of the article goes on to talk about insurance rates, unemployment, cancer costs, and the economic hardship of the young.  I’ve scanned the research and it seems that they used self-reports from the subjects to determine why they were forgoing care.

That leads me to ask: is it likely that cost is the only reason for the 67% difference in care-seeking?  Perhaps the dominant paradigm that heath care is too expensive provides a convenient answer without digging into more complex and uncomfortable reasons.  Purely speculating, I’d wonder about emotional exhaustion of Damocles Syndrome, a feeling of invulnerability from both the level of mental maturity and having survived cancer already, resentment at time lost to illness and a desire to just move on until a symptom appears,  an avoidance of potential bad news, for the sake of themselves and loved ones, or any of countless personal or cultural reasons.

Maybe I’m completely wrong on this, but it seems like the researchers put a lot of thought into economic factors and few into human ones.  It’s not that the study is inaccurate — it reports what the subjects told them — but that the conclusions seem superficial.