Table 2.

Predictive power of parental beliefs about child development for children’s skills.

	Beliefs on impact of parents’ inputs at:
	13–16m		19–22m		25–28m		37–40m
ASQ-SE at
19–22m	0.19*	4.74%	0.17*	3.87%
	(0.10)	68	(0.10)	77
37–40m	0.14	0.99%	0.29*	4.23%	0.31*	3.86%	0.40**	7.76%
	(0.18)	67	(0.16)	74	(0.18)	73	(0.16)	76
PLS at
13–16m	0.20**	5.14%
	(0.10)	83
19–22m	0.84***	13.15%	0.99***	15.35%
	(0.27)	68	(0.27)	77
25–28m	1.05***	13.24%	1.24***	15.69%	1.09***	9.82%
	(0.33)	68	(0.34)	76	(0.39)	75
31–34m	0.94***	12.22%	1.41***	18.19%	1.35***	13.60%
	(0.31)	69	(0.35)	76	(0.40)	75
37–40m	1.24***	14.40%	1.40***	18.65%	1.40***	15.03%	1.25***	14.09%
	(0.38)	64	(0.35)	72	(0.40)	70	(0.37)	73
PPVT at
37–40m	0.32***	14.02%	0.35***	15.98%	0.35***	12.67%	0.31***	11.66%
	(0.10)	63	(0.10)	71	(0.11)	69	(0.10)	72

The rows correspond to children’s skills measured at five different time points. ASQ-SE stands for Ages and Stages Questionnaire: Social Emotional, PLS stands for Preschool Language Scale, and PPVT stands for Peabody Picture Vocabulary Test (see Supplementary Information, section 4.3). They are regressed on parents’ beliefs measured at four different time points, in separate simple linear regressions. Both the child outcome variables and the beliefs variables are standardized. For each regression, the table gives the coefficient of the beliefs measure, its standard error in parentheses below, the R-squared in italics next to the coefficient, and the number of observations below the R-squared. *For p-value < 0.1, **for p-value < 0.05, ***for p-value < 0.01. p-values are based on two-sided Student’s t tests without adjustment for multiple comparisons.