Summary
A view holds that expertise level depends on practice alone and that certain types of practice are important or unimportant. Supporting evidence largely comes from studies using a correlational retrospective recall paradigm, usually with small samples. Initially, these studies were partially replicated with 533 international chess players. Log number of games played was the strongest predictor of latest performance rating. Then, effects of study hours, having had coaching and the number of games played were examined longitudinally to control for key variables confounded in the retrospective recall paradigm. Groups with a nearly 5β1 median difference in weekly study hours, roughly equated on time in the domain and the number of games played, were observed over 7βyears. More study hours had negligible impact. Coaching had some effect over time, and the number of games had a strong effect even when participants were equated on time in the domain. Previous studies show that a factor other than the number of games is important in developing chess expertise. Study is a weak factor at best and could not be that important factor. Chess expertise apparently does not depend on practice (study and the number of games) alone. Copyright Β© 2011 John Wiley & Sons, Ltd.