Skip to main content
. Author manuscript; available in PMC: 2010 May 7.
Published in final edited form as: Am Econ J Appl Econ. 2009 Oct 1;1(4):34–68. doi: 10.1257/app.1.4.34

Table 3.

Robustness Tests of Random Assignment

Panel A: Modified Randomization Test
Dependent Variable,X =
Ability (1) Driving Distance (2) Putts per Round (3) Greens per Round (4) Years of Experience (5) Length of Name (6)

Avg(X ), playing partners −0.017 (0.015) 0.002 (0.015) −0.002 (0.012) −0.004 (0.014) −0.010 (0.012) −0.013 (0.014)
Leave-me-out Avg( X ), urn −10.803 (1.629) −10.189 (1.210) −12.529 (1.653) −12.033 (2.665) −15.793 (3.795) −14.587 (1.364)
Tournament × Category F.E. Y Y Y Y Y Y
Leave-me-out urn mean included Y Y Y Y Y Y
R2 0.542 0.632 0.579 0.580 0.593 0.515
N 8801 8646 8646 8646 8646 8646
Panel B: Biased Randomization Test
Dependent Variable,X =
Ability (1) Driving Distance (2) Putts per Round (3) Greens per Round (4) Years of Experience (5) Length of Name (6)

Avg(X ), playing partners −0.087 (0.022) −0.081 (0.021) −0.071 (0.022) −0.080 (0.023) −0.050 (0.020) −0.063 (0.020)
Tournament × Category F.E. Y Y Y Y Y Y
Leave-me-out urn mean included N N N N N N
R2 0.254 0.382 0.163 0.251 0.167 0.035
N 8801 8646 8646 8646 8646 8646

Notes:

a

Results from estimating equation (2).

b

Avg(Ability) is the average handicap of a player’s playing partners, as described in the text.

c

All specifications in Panel A include the average of all of the other players in your tournament-by-category urn (not including yourself). This control is necessary to produce a well-behaved randomization test.

d

Standard errors are in parentheses and are clustered by playing group.

e

All regressions are unweighted (unlike the specifications in the other tables), since the weighting is not appropriate for the randomization test.