Truth be told, I looked at the numbers for Lewis and Walker, and my first thought was that there wasn't a significant difference between them. But, I ran a quick test, and proved myself wrong:
The FREQ Procedure
Table of make by player
make player
Frequency‚
Percent ‚
Row Pct ‚
Col Pct ‚lewis ‚walker ‚ Total
---------ˆ--------ˆ--------ˆ
0 ‚ 661 ‚ 805 ‚ 1466
‚ 28.07 ‚ 34.18 ‚ 62.25
‚ 45.09 ‚ 54.91 ‚
‚ 59.71 ‚ 64.50 ‚
---------ˆ--------ˆ--------ˆ
1 ‚ 446 ‚ 443 ‚ 889
‚ 18.94 ‚ 18.81 ‚ 37.75
‚ 50.17 ‚ 49.83 ‚
‚ 40.29 ‚ 35.50 ‚
---------ˆ--------ˆ--------ˆ
Total 1107 1248 2355
47.01 52.99 100.00
Statistics for Table of make by player
Statistic DF Value Prob
------------------------------------------------------
Chi-Square 1 5.7332 0.0166
Likelihood Ratio Chi-Square 1 5.7305 0.0167
Continuity Adj. Chi-Square 1 5.5311 0.0187
Mantel-Haenszel Chi-Square 1 5.7308 0.0167
Phi Coefficient -0.0493
Contingency Coefficient 0.0493
Cramer's V -0.0493
Fisher's Exact Test
----------------------------------
Cell (1,1) Frequency (F) 661
Left-sided Pr = F 0.9926
Table Probability (P) 0.0019
Two-sided Pr <= P 0.0171
Sample Size = 2355

]]>