I find it disappointing that Slate has included this story at all. These findings merely indicate a correlation, and they say nothing about causation. We could just as likely attribute greater skill at sprinting to having dark skin rather than having active ACTN3. Or we could argue that the Jamaican diet explains these differences rather than genes (the logic is the same).

In order to establish causation, we would need to select a group of roughly equally skilled athletes who all lack this gene, genetically modify half of them so that this select subgroup subsequently had an active ACTN3 gene, and then test the two groups to see if we see any differences in their athletic performances. If we notice such a difference, then we might be able to make a causal link between these variables. Of course such research would violate a number of ethical principles; however, for the sake of argument, we would need a procedure like this in order to begin to consider the questions posed by the article. I wish the Slate author had discussed this significant weakness in her paper. Without such a discussion, too much room is left for misinterpretation of the summarized findings.

Another weakness of this article is the fact that it does not appear that any of the winning Jamaican athletes referenced underwent any testing to see if they actually have this genetic trait. So, again, we don't know if the trait had anything to do with their success at all -- remember, not all Jamaicans have this trait. And for those who managed to win without this trait, how do we explain this?

This piece really annoys me for two reasons. First, it undermines the legitimacy of the hard work that these Jamaican athletes have put in because it implies that they had an unfair advantage that non-Jamaican athletes lack. Second, it does not address the history of racism behind such research. Specifically, I have yet to see a study examining the genetic cause of why White athletes tend to dominate at water polo, swimming and diving, and other stereotypically "White" sports. There is a reason that these questions are not asked, and I think addressing this issues would be at least as (if not more) interesting as the questions posed by the present article.

In the future, I hope that Slate does a better job distinguishing between correlational designs and true experiments, as well as addressing how the biases of researchers influences not only the questions they ask, but the conclusions they draw concerning their findings.

Slate really dropped the ball on this.

It's an old (and racist) saw to attribute athlete ability to genetic factors: it's like saying Canadians have a protein that makes them great hockey players...More to the point with Jamaica's success is the lack of out-of-competition dope testing: anyone that thinks a 9.69 100 or a sweep of the Women's podium comes from "better training" is hopelessly naive. "Otherwordly" performances in sports where records are usually rarely broken, and when broken, in tiny increments are inevitably through PED's. Remember, Marian Jones never tested positive, but she was on the full program.

On January 16, 1988, Jimmy "The Greek" Snyder was fired by the CBS network (where he was a contributor to the NFL Today program since 1976) after commenting to WRC-TV reporter Ed Hotaling in a Washington, D.C. restaurant that African Americans were naturally superior athletes at least in part because they had been bred to produce stronger offspring during slavery: "The black is a better athlete to begin with because he's been bred to be that way — because of his high thighs and big thighs that goes up into his back, and they can jump higher and run faster because of their bigger thighs. This goes back all the way to the Civil War when during the slave trading, the owner — the slave owner would breed his big black to his big woman so that he could have a big black kid." Snyder died of a heart attack in Las Vegas, at the age of 76. He is buried at the historical Union Cemetery in Steubenville, Ohio.

Windorchard, I could not have said it better! The journalist behind this article exemplifies the dangerous disconnect between true science and science interpretation and reporting. The average layman will read that article and believe it to be a legitimate representation of the actual science involved -- when it is not.

I also like your comment about how the biases of scientist affect the type of questions they formulate. I have been a firsthand witness to this, and can say that science becomes repulsive when its methodology is skewed by social stereotypes.

Seriously, reporters need a course in causal inference!

reporters are mostly just lazy chumps who watch a battle from a safe hillside and, when it is over, come down and shoot the wounded.

Since the Olympics started, has Slate published any article that actually praised the athlete without somehow trying to downplay their accomplishments by giving reasons why their winning was no big deal because.... ?

Slate needs to stay away from trying to cover sports related topics.

Speaking of "old saws", its an old saw to point out that correlation is not causation. However, people that enjoy pointing that out frequently don't know the actual relationship between correlation and causation.

For instance, the statement:

"These findings merely indicate a correlation, and they say nothing about causation."

is simply wrong. Correlation DOES say something about causation. Correlation is a necessary condition for causation, although not a sufficient one. Thus establishing a strong correlation is suggestive of causation, but does not actually prove causation.

A great many well executed research programs rely on demonstrating correlation, in cases where establishing causation in a controlled experiment would be difficult or impossible. To suggest that this is bad science, or not science, is to be ignorant of how science is practiced. That being said, it is of course important for researchers to not over-reach in stating the implications of findings of correlation.

the fact that there is an history of racism behind ethnicity-based research does NOT mean that we will never be able to conduct any research, or postulate theories that will have to be proved or disproved. Your demand for totally outrageous conditions for performing research (genetically modify people …) thinly masks the fact you do not want ANY research that might find a difference between ethnic groups. BTW I am not claiming that such difference exists or was proved in this case; but I am annoyed at the attempt of muzzling any discussion of these subjects by raising the "racism" flag. At some point I believe this is self-defeating, and will contribute to more people believing in this.

as for this:

"Specifically, I have yet to see a study examining the genetic cause of why White athletes tend to dominate at water polo, swimming and diving, and other stereotypically "White" sports"

Very weak, my friend. All these sports, as has been pointed time and again, need MONEY which does not exist in poor, non-white countries, nor in poor, non-white population areas in developed countries.

There should be a study done on this. They should take all the finalists in the men's and women's 100m races and do double blind genetic tests on all the white runners and on all the black runners.

Training and traveling to international track meets is not a "money sport"? Pretty weak arguement. I'm pretty sure they have plenty of water in Jamaca to swim in.

funny thing is, if there was a genetic study that proved that white people where genetically better suited to swimming. it would be interpreted by most whites as a mark of superiority. while when whites report black physical prowess it is perceived as denigrating.

Pretty racist statement, rustedspear.

"Correlation DOES say something about causation. Correlation is a necessary condition for causation, although not a sufficient one. Thus establishing a strong correlation is suggestive of causation, but does not actually prove causation"

It depends. Correlation doesn't ALWAYS say something about causation. Because two variables can be correlated just by chance alone, and in fact be completely unrelated.

Correlation speaks more to association rather than causation, and even that association can be completely spurious or driven by something else. Also saying that strong correlation is suggestive of causation is extremely sloppy science.

I do agree that correlation is necessary but not sufficient criterion. Correlation is a very small slice of the pie and we should not let it carry the burden of proof.

Correlation says nothing about causation; however, it does, I agree, suggest some likely candidates. The cliché (correlation is not causation) exists because it it apt (not because the uniformed like to use it to bully the even less informed). You cannot with mere correlation remove the possibility that the relationship observed between any two variables is not spurious without an experiment. No scientific field that I am aware of establishes causation through correlation. One might explore possible causal agents, but one does not establish causation this way. But hey...maybe you know more about this than I and there is a technique you know for demonstrating causation from correlation. Please give me the reference and I will change my song if it seems sound.

As I was taught to practice science (and perhaps it differs across fields), one does not "prove" anything with science; one supports hypotheses. So, science is not an objective enterprise, it is one based on aggregated studies, analyses, and opinions. If science was clear and objective, we would not be having this debate.

I was taught that one must ask why one is performing a particular study and also consider what the ramifications of your findings might be. And yes, I was taught that some studies perhaps should not be done. I suspect you would agree that there are some studies that should not be done. We merely appear to disagree on which studies fall into which category.

My concern with cross-cultural studies like the one that precipitated this discussion lies in the fundamental understanding of correlation, causation, and research methods. Unless you can actually manipulate your independent variables, it is impossible to make causal statements. Therefore, all comparisons of men and women, Blacks and Whites, or any group of people who already come to a study having already been ascribed a quality that a researcher might like to use as an independent variable is automatically confounded. This is not something that is my opinions. This is what the scientific method says.

Does this mean we ought never conduct such studies. Certainly we ought! Sometimes such studies are very useful. However, those of us who do such work (and I sometimes do such work) must be conscientious to convey to those who read our work that our findings are merely correlational, and therefore must be carefully considered.

Further, I believe that researchers have a responsibility to try (even if we are not always successful) to not do research that harms people, particularly the people we are studying. In the present era, many cross-cultural studies (though not all of them) have great potential for harm, and must be undertaken with great thoughtfulness.