Table 14.

Children's Interactions and Learning Behaviors by Condition

	Tools (n = 32)		Control (n = 28)		Test of Condition Effects
	M	SD	M	SD	B	SE	d
Hypothesis 3. Children's Interaction
Associative Interactions	0.104	.039	0.068	.028	.035*	.010	1.054
Cooperative Interactions	0.021	.018	0.006	.008	.015*	.004	1.081
Parallel Interactions	0.470	.051	0.479	.053	−.014	.015	−0.175
Hypothesis 4. Children's Learning Behaviors
Overall Level of Involvementa	2.379	.167	2.326	.201	.045	.048	0.293
Large Group Level of Involvementa	2.692	.179	2.571	.201	.121*	.050	0.545
Centers Level of Involvementa	2.924	.166	2.834	.170	.090*	.042	0.649
Unoccupied	0.047	.018	0.045	.023	.001	.005	0.100
Disruptive	0.009	.006	0.007	.006	.003	.001	0.333

Note. Tools of the Mind was the reference group in each model as such positive coefficient indicate Tools classrooms had higher rates of an observed variable compared with the control classrooms. Coefficients reported are unstandardized regression coefficients from multilevel regression models. Estimates of effect sizes of curriculum differences provided are Cohen's d estimates standardized mean difference. All variables captured using the Child Observation Protocol. Unless noted, estimates are the proportion of sweeps across the three observations a given behavior was observed.

SD = standard deviation; SE = standard error.

^aLevel of involvement was scores on a 1–5 rating scale.

^* p < .05.