Abstract des Autors

Indices of team performance for the 64 teams competing in the 1991 National Collegiate Athletic Association (NCAA) Division 1 Men's Basketball Tournament were analyzed for predictive utility using two full-model regression analyses. Values for six team indices were used as predictor variables (points per game, points allowed, number of field goals, number of free throws, number of three-point goals, number of rebounds); USA Today/CNN and UPI final season cumulative point total rankings were used as the two criterion variables. Results of analyses revealed different patterns of predictors for both dependent variable scales. The variable points per game emerged as the most salient predictor for both equations. No other variable emerged as a substantive predictor for the USA Today/CNN scale. For UPI rankings, points per game was joined by number of field goals and number of rebounds as the three major contributors. Verf.-Referat