Table 3.

Affect recognition and memory for faces as predictors of playground behavior

Behavior	Memory for Faces Delayed				Memory for Faces Immediate				Affect Recognition				Regression Model
	t	β	p	Boot p	t	β	p	Boot p	t	β	p	Boot p	F	p	R2
Self-Play	−3.01	−0.61	0.01**	0.01**	1.76	0.33	0.09	0.05	0.07	0.11	0.94	0.94	3.01	0.02*	0.24
Coop Play	2.98	0.59	0.01**	0.01**	−1.46	−0.27	0.15	0.11	−0.19	−0.03	0.85	0.85	3.26	0.01**	0.25
Equip Play A	−3.35	−0.68	0.02*	0.02*	1.95	0.37	0.06	0.04	0.92	0.14	0.36	0.31	2.60	0.04*	0.21
Equip Play G	2.48	0.52	0.02	0.02*	−1.05	−0.20	0.30	0.38	0.25	0.04	0.80	0.79	2.17	0.07	0.18
Verbal Bout	1.25	0.26	0.22	0.25	−0.85	−0.16	0.39	0.41	−1.01	−0.16	0.32	0.14	2.32	0.06	0.19

Results from primary aims are presented. Due to violation of normal distribution, bootstrap resampling was used and results are presented based on 1,000 samples (Boot p). For each significant predictor in the regression equation to control for multiplicity effects across the five behavioral variables, we used the Bonferroni-Holm method, a step down procedure in which a sequential set of tests are conducted that involve layered adjustments of alpha levels based on the number of elements remaining in the original set.

t t statistic, β beta value, p corrected alpha levels using Bonferroni-Holm method (*p < 0.05, **p < 0.01), Coop cooperative, Equip equipment, A alone, G group.