On the relationship between objective measures and performance in basketball: Selecting teams of seventh grade girls

The present study examines the utility of several objective measures to predict which girls were selected for one of six junior high school basketball teams. One purpose of this research was to examine the extent to which differences between coaches’ selection strategies might have contributed to the apparently contradictory results among previous attempts to identify correlates of coaches’ ratings or team selections. Linear discriminant models of six coaches’ team selections were constructed from measures of basketball and athletic skills, strength, physical measures, and competition anxiety obtained during the first week of practice. These models provided greater-than-chance agreement with coaches’ player selection for each team (87%—93% correct classification). A cross-modeling procedure revealed that the six coaches’ selections could be modeled in terms of one of three approaches to selection: A) Select the aspirants who possess the best basketball skills (three teams), b) select the aspirants with greatest size and strength and least competition anxiety (two teams), and c) select only the aspirants with both types of attributes (one team). The results indicate that it may be difficult to identify a single set of measures to select skillful or potentially skillful basketball players, because of differences in how coaches might choose to define these terms.