Table 1.

Mixed-effects logistic regression models on the odds of children learning through a specific learning process.

	Model 1: imitate		Model 2: observe		Model 3: practice		Model 4: teach		Model 5: play-practice
	Odds ratio (CI lower, CI upper)	P-value	Odds ratio (CI lower, CI upper)	P-value	Odds ratio (CI lower, CI upper)	P-value	Odds ratio (CI lower, CI upper)	P-value	Odds ratio (CI lower, CI upper)	P-value
Age (1-year increase)	0.53 (0.34, 0.84)	<0.01	0.7 (0.57, 0.88)	<0.01	1.03 (0.88, 1.2)	0.73	0.85 (0.63, 1.17)	0.32	1.42 (1.12, 1.8)	<0.01
Sex (male)	0.39 (0.02, 6.94)	0.52	0.21 (0.04, 1.26)	0.09	0.31 (0.09, 1.09)	0.07	3.25 (0.29, 36.06)	0.34	5.73 (0.91, 35.97)	0.06
AIC	111.8		284.1		428.5		183		503.3
Log likelihood	−51.9		−138.0		−210.2		−87.5		−247.6

Odds of children using imitative learning (model 1), observational learning (model 2), practicing a skill (model 3), being taught (model 4), and play-practicing (model 5) with respect to sex and age. All models control for the focal child’s ID. N = 449 entries for 34 focal children. Data on 6 participants were not included in the models because we did not have age estimates for those participants.