Table 2.

Coefficients of second-stage Heckman GEEs regressions predicting strategy quality

Variable	Model 1		Model 2		Model 3
	Coeff.	p	Coeff.	p	Coeff.	p
Cultural Diversity			0.014	0.677	0.031	0.398
			(0.035)		(0.036)
Cultural Diversity × Rule Change Time					− 0.003	0.002
					(0.001)
Cultural Diversity × Number of Virtual Characters					0.008	0.043
					(0.004)
Cultural Diversity × Home Court					0.153	0.029
					(0.070)
Rule Change Time	− 0.002	0.000	− 0.002	0.000	− 0.002	0.000
	(0.000)		(0.000)		(0.000)
Number of Virtual Characters	− 0.001	0.284	− 0.001	0.283	− 0.001	0.314
	(0.001)		(0.001)		(0.001)
Home Court	0.053	0.000	0.054	0.000	0.045	0.000
	(0.011)		(0.011)		(0.013)
Team Harmony	0.442	0.000	0.443	0.000	0.442	0.000
	(0.082)		(0.082)		(0.082)
Losing Game	− 0.013	0.073	− 0.013	0.072	− 0.013	0.060
	(0.007)		(0.007)		(0.007)
BO1	− 0.057	0.000	− 0.057	0.000	− 0.066	0.000
	(0.016)		(0.016)		(0.017)
BO3	− 0.028	0.034	− 0.027	0.038	− 0.027	0.043
	(0.013)		(0.013)		(0.013)
TFG	− 0.002	0.218	− 0.002	0.193	− 0.003	0.074
	(0.002)		(0.002)		(0.002)
Team Average SIS	0.002	0.537	0.001	0.672	0.001	0.845
	(0.003)		(0.003)		(0.004)
Professional Age	− 0.000	0.872	− 0.000	0.844	− 0.000	0.863
	(0.001)		(0.001)		(0.001)
Member Turnover	− 0.001	0.485	− 0.001	0.472	− 0.001	0.634
	(0.002)		(0.002)		(0.002)
Team Familiarity	0.000	0.492	0.000	0.478	0.000	0.488
	(0.000)		(0.000)		(0.000)
Coaching Experience	0.003	0.166	0.003	0.217	0.003	0.176
	(0.002)		(0.002)		(0.002)
Same-country Coach	0.011	0.155	0.011	0.150	0.012	0.143
	(0.008)		(0.008)		(0.008)
IMR	− 0.017	0.041	− 0.017	0.040	− 0.017	0.041
	(0.008)		(0.008)		(0.008)
Constant	1.112	0.000	1.112	0.000	1.113	0.000
	(0.009)		(0.009)		(0.009)
Pseudo R2	0.051		0.051		0.053

Year dummies are included in all models. Standard errors are in parentheses; actual p values are reported. All tests are two-tailed. N = 8070. IMR is the inverse Mills ratio from the first stage. Since GEE is not a likelihood-based method, pseudo R² are reported as recommended by Chatterjee and Hambrick (2007) and Grijalva et al. (2020).