Practice Doesn’t Necessarily Make Perfect, and Neither Is The 10,000 Hour Rule
By Kyle Hill on August 21, 2014
What makes someone truly great at what they do? How did The Beatles take the world by storm and how did Bill Gates so drastically change the technological landscape? Well, if you’ve read Malcolm Gladwell’s Outliers or the numerous others citing the work of a certain Swedish psychologist, the answer is practice. More specifically, there seems to be an odd clustering of experts who have engaged in deliberate practice for over 10,000 hours. Practice, as they say, makes perfect.
But the first formal analysis of the link between deliberate practice and human performance says otherwise. Practice is certainly important, but far less than other factors. The “10,000 hour rule” has bit the dust.
In 1993, Swedish psychologist K. Anders Ericsson and his team published a study looking at how deliberate practice helped people become experts. Deliberate practice is practice with the intent of improvement and focusing on weaknesses — the difference between playing a game of pick-up basketball with friends and shooting 1,000 free-throw shots. Ericsson and his team concluded that, “We view elite performance as the product of a decade or more of maximal efforts to improve performance in a domain through an optimal distribution of deliberate practice.” That mouthful of academia is what Gladwell based sections of Outliers on, and thus the 10,000 hour rule was born.
It’s an equalizing thought — apart from great genes, only more practice holds you back from being a Tiger Woods or a Michael Jordan. But that idea has gotten a lot of push-back from researchers in the 20 years since the Ericsson study, and now the largest study to date on the same topic has found that deliberate practice explains far less than what the 10,000 hour rule claims.
Published in Psychological Science this July, the new study from Brooke Macnamara, David Hambrick, and Frederick Oswald is a meta-analysis of 88 different studies covering human performance and practice across music, games, sports, professions, and education. A meta-analysis is like a study of studies. It sets up criteria for what a good study should look like, searches for dozens of studies that could be included, and then weeds out the studies that aren’t up to par. By looking at the data from dozens of studies all put together, scientists can sometimes pull out larger trends in a field that any one study might not have the statistical strength to address. Macnamara’s study covers over 11,000 human participants.
So, when they set up all the variables, consider all the caveats, and group the data that belong together, what did the meta-analysis find? Well, in short, that deliberate practice isn’t even half of the reason how experts are made. Take a look at a figure from the (open access) study below:
When you are hunting for correlations between variables, what you are really doing is trying to find out how much the variable you are looking at explains the correlation you find — how much of the variance is explained. Since correlations range from 0.0 to 1.0, that means that if reading twice a week had a 0.5 correlation with IQ scores, for example, then reading twice a week only explains 25% of everything that could account for a higher IQ (it’s only 25% because correlations are mathematically squared, so to get to the variance you have to use 0.5*0.5). Averaging out all the correlations between deliberate practice and performance in the studies Macnamara included in the meta-analysis, the correlation was 0.35.
This means that, overall, practice accounted for just 12% of all the stuff which might explain expert-level performance. More often than not, there are variables besides practice that are driving levels of performance. The figure above shows the variance explained for each domain.
10,000 hours of practice may not make perfect, but that does not mean it is useless. Practice will help you get better at some task, of course, but not nearly as much as the 10,000 hour meme suggests. The fascinating part is that because so little of performance is explained by practice, there are still unknown variables out there that are just waiting to be discovered and capitalized on. Maybe it’s good genetics, or socio-economic opportunity, or maybe it’s just plain old luck. But whatever they are, they aren’t going to be as simple and clear cut. Life is complicated.
That’s all the more reason to work at and stay passionate about what you love, because science.
Kyle Hill is the Chief Science Officer of the Nerdist enterprise. Follow the continued nerdery on Twitter @Sci_Phile.