Mean Performance Measure
T Test of Difference in Means
.0903 .0259 .4729 .3969 .1853
0.102 –1.386 –2.394** 0.703 –1.160
Assists Blocks Field-goal percentage Points Rebounds
.0915 .0193 .4514 .4082 .1681
Burdekin et al. / NBA FANS INDIFFERENT TO RACE?
Performance Statistics for NBA Starters and Bench Players
T Test of
Difference in Means
2. Bench Players
Assists Blocks Field goals perce
NOTE: This table shows the mean values of five key performance measures averaged over three seasons (1996-1998) that the player played as a starter, or bench player, respectively. A player is classified as a starter if he is one of the five players on the team who played the most total minutes over the course of the season. A player is classified as a bench player if he is on the team’s roster but is not one of the five players on the team who played the most total minutes over the course of the season. Except for field-goal per- centage, all performance measures are standardized by total minutes played in the season. Data are from the NBA Web site (see www.nba.com). In section 1, total sample size equals 228 (35 starters who are White and 193 starters who are Black). In section 2, total sample size equals 476 (110 bench players who are White and 366 who are Black). **Significant at the .05 level, two-tailed test.
through 1998-1999 seasons. We measure team racial composition in three ways: the percentage of all team members who are White, the percentage of a team’s bench players who are White, and the percentage of a team’s starting players who are White. Customer discrimination implies a positive relationship between the team racial composition and the metro-area racial composition. Note that although we do not differentiate between American and foreign-born players, any reduced interest by American White fans in watching White European players would actu- ally bias downward the implied importance of race in player allocation and atten- dance outcomes—and would, if anything, understate the value attached to U.S.–born White NBA players.6
We would, however, expect team racial composition to be more important in smaller metropolitan markets because such markets have fewer potential custom- ers to fill an arena’s seats. Assuming that some fraction of local customers pos-