Here's a sobering feature story about a former college football player, Ryan Hoffman. Ryan used to be an offensive lineman for the University of North Carolina football team. Now he is living in Florida, homeless, jobless, suffering from memory problems.
You may have seen recent news stories written about a suspected link between football experience, especially game-induced concussions, with later dementia, cognitive problems, depression, or drug use. Most such stories note that, scientifically speaking, it may be difficult to show a causal link between playing football and having these kinds of problems.
The Times story about Ryan Hoffman states very clearly the "compared to what" problem that is introduced in Chapter 2.
For example, the journalist writes:
Just like their N.F.L. counterparts, Hoffman and those former college players have been left to wonder: Did football do this? Are the hits to the head I took the reason for my decline? Or would I be in this condition even if I’d never played a down?
They might never know the answer, because a definitive answer might not exist.
Hoffman blames football for scrambling his brain, but at this point it is impossible to disentangle what could be football-related brain injuries from his subsequent drug use and possibly genetic mental illness. He simply cannot be sure. No one can.
The football-brain injury link is a good illustration of the difficulty of basing our beliefs on experience, compared to empirical evidence. Former players who suffer from depression, cognitive impairment, or memory loss have no way of knowing if they would have had the same symptoms if they had never played football.
a) Recall that you can make a 2×2 table comparing two possible "treatments" and two possible "outcomes." In Chapter 2, the tables compare using Dr. Rush's bleeding treatment (or not) to the outcomes of recovery (or not). We'd need a similar table to establish a definitive link between football and brain injuries. Sketch such a table.
b) What kind of study might you conduct to definitively test for a connection between football and brain injury? Would such a study be ethically and practically possible to do?
c) Chapter 13 introduces quasi-experimental designs. Since we cannot, ethically or practically, randomly assign people to play football or not, how might we use a quasi-experimental design to test for a connection between football and brain injury?