Submitted to the NY Times, Oct 18, 2011
The idea that chess will improve academics has made it into the New York Times ("Maybe teach them math, science and chess," Oct 15).
The Times article presents no evidence that chess is helpful. To my knowledge, there is only one study testing this idea: "The Effect of Chess on Reading Scores: District Nine Chess Program Second Year Report" was published by the American Chess Foundation (!!), and claimed that 53 students who voluntarily participated in a chess program in New York improved five percentiles in reading over the year (from the 58th percentile to the 63rd).
Nearly all the gains, however, were from six children who made unbelievable improvements, ranging from 38 to 66 percentiles. If we remove these outliers, the average gain for all 53 students is not impressive, less than two percentiles.
The case for chess, in other words, depends on unusual gains made by six children in one study done about 20 years ago.