Table 3.

Mean Average Accuracy and hypothesis tests

Test Variable: Average Accuracy	H1 test		H2 test		H3 test
	Males (n=235)	Females (n=230)	Skill-Game (non-problem) Gamblers (n=99)	NonSkill-Game (non-problem) Gamblers (n=30)	Problem Gamblers (n-77)	Non-Problem Gamblers (n=129)
mean	.727	.709	.689	.690	.693	.689
t-stat (p-value)	1.2695 (p=.205)		.021 (p=.984)		-.160 (p=.873)
Exploratory--comparison of Average Accuracy by current gambling frequency
		Gfreq=0 (n=259)	Gfreq=1 (n=93)	Gfreq=2 (n=62)	Gfreq=3 (n=43)	Gfreq=4 (n=8)
	Mean	.740	.716	.674	.670	.634
	St. dev	.144	.158	.163	.166	(.211)

SkillGame gambler t-test used the subset of participants who reported current gambling frequency greater than zero. A similar test did not reveal any statistically significant difference in Accuracy between nonproblem gamblers who exclusively played skill-games compared to those who played a mix of skill games and games of chance, but our data set only includes n=12 current gambler participants who reported exclusively playing gambling games of skill. “Gfreq” = Gambling Frequency ∈ [0,4], which describes self-reported current gambling frequency (0,1,2,3,4 indicates responses of “never”, “less than once a month”, “once or twice a month”, “once or twice a week”, or “daily”. Nongambler participants were scored as Gfreq = 0