Saturday, July 11, 2015

Correlation, Cause, Tomato, Tomahto

Seriously, I'd like nothing better than to find concrete proof that sitting through my classes will improve my students' health and longevity.

But this...

A new study by researchers at the University of Colorado Denver, New York University, and the University of North Carolina at Chapel Hill estimates the number of deaths that can be linked to differences in education, and finds that variation in the risk of death across education levels has widened considerably. The findings, published July 8 in the journal PLOS ONE, suggest that lacking education may be as deadly as being a current rather than former smoker. 
This isn't just headline hyperbole. The actual study in PLOS-one is titled

Mortality Attributable to Low Levels of Education in the United States

You're going to feel a pinch...

Have we all decided that "associated with" is now the same thing as "attributed to?" [1]

BRB, off to cure lung cancer and emphysema with my tooth-whitening kit. [2]

[1] The authors do cite  this article to support their claim of a causal relationship between education and health. The article itself, however, acknowledges the causal inference problem,  states, "We are unaware of randomized experiments of education at older ages [than preschool]," and concludes that "the identification of causality has proved challenging in studies linking education and income to health...A better understanding of the sorts of questions raised in this essay remains the task for the next generation of research on SES and health."

[2] Am I just going to snark, or do I have any actual constructive suggestions? Well both. Here's my idea: follow up on the participants in CUNY's ASAP program, who were randomly selected to receive a suite of academic and financial supports, and ended up completing their degrees at nearly twice the rate of the control group. 


  1. While the general public never have been great with statistical thinking, I'm convinced that media coverage of climate change has obliterated any they ever did have.

    1. The next time you hear, "The average lifetime earnings of an average college graduate are $1M higher than the average lifetime earnings of an average high-school graduate who did not attend college," yell BINGO!

    2. *Thank* you. And those extra lifetime earnings totally appeared out of nowhere, high-paying jobs will materialize to accommodate all the extra college grads, and the differential has nothing to do with the fact that the profusion of degreed graduates has completely devalued the high school diploma.

  2. So when my students tell me they will face dire consequences if I don't give them a passing grade, I should believe them?