Data Mining in the Exploration of Stressors Among NCAA Student Athletes.

Collegiate student athletes face psychological stressors in adjusting to campus life. This study used preexisting, nationally representative data administered by the National Collegiate Athletic Association for student athletes in 2010 to explore the conjunctive relationships among demographics, personal characteristics, social contexts, and physical condition in predicting perceived stress. The number of valid samples was 19,967 from 609 member institutions. A data mining methodology (i.e., SEARCH) was applied to model the distribution of the perceived stress. Results showed that significant stressors included the variables related to academics, physical well-being, and social contexts. Academic anxiety was the most important predictor, and its interactions with abusive coaching behavior and an inclusive team environment were shown to reduce perceived stress. Sufficient sleep was also found as a moderator in the positive relationship between perceived stress and academic anxiety.