What determines adult cognitive skills? Influences of pre-school, school, and post-school experiences in Guatemala

Behrman, Jere R.; Hoddinott, John; Maluccio, John A.; Soler-Hampejsek, Erica; Behrman, Emily L.; Martorell, Reynaldo; Ramírez-Zea, Manuel; Stein, Aryeh D.

doi:10.1007/s40503-014-0004-4

Latin American Economic Review

Table 3 Estimated impacts of pre-school, school, and post-school experiences on cognitive skills (N = 1,448)

From: What determines adult cognitive skills? Influences of pre-school, school, and post-school experiences in Guatemala

Life-cycle stage	Representation	Set 1A		Set 1B		Set 2		Set 3
Life-cycle stage	Representation	OLS	IV-GMM	OLS	IV-GMM	OLS	IV-GMM	OLS	IV-GMM	LIML
	Panel A: Verbal Z scores
E₁	HAZ score at age 6	0.202 (0.028)	−0.071 (0.146)			0.073 (0.021)	0.107 (0.103)	0.073 (0.021)	0.184 (0.116)	0.089 (0.181)
E₂	School attainment			0.221 (0.006)	0.101 (0.018)	0.218 (0.006)	0.105 (0.018)	0.218 (0.006)	0.087 (0.020)	0.070 (0.028)
E₃	Skilled job tenure							−0.001 (0.004)	0.028 (0.014)	0.029 (0.017)
	Age at interview in 2002–2004	0.220 (0.094)	0.058 (0.119)	0.076 (0.075)	0.097 (0.054)	0.111 (0.074)	0.173 (0.087)	0.111 (0.074)	0.201 (0.095)	0.109 (0.126)
	Age at interview squared × 10	−0.037 (0.014)	−0.014 (0.018)	−0.012 (0.012)	−0.018 (0.009)	−0.018 (0.011)	−0.029 (0.013)	−0.018 (0.011)	−0.034 (0.014)	−0.020 (0.019)
	Constant	−2.690 (1.501)	−0.520 (1.710)	−2.164 (1.201)	−1.709 (0.836)	−2.574 (1.17)	−2.781 (1.267)	−2.585 (1.177)	−2.993 (1.372)	−1.582 (1.803)
	Test diagnostics
	F Statistic	26.5	17.4	472.2	32.3	401.7	23.9	335.6	18.9	9.6
	[p value]	[< 0.01]	[< 0.01]	[< 0.01]	[< 0.01]	[< 0.01]	[< 0.01]	[< 0.01]	[< 0.01]	[< 0.01]
	Hansen J test [p value]		[0.03]		[0.17]		[0.12]		[0.16]	[0.19]
	Hausman test [p value]		[0.32]		[< 0.01]		[< 0.01]		[< 0.01]
	Panel B: Nonverbal Z scores
E₁	HAZ score at age 6	0.192 (0.034)	0.101 (0.128)			0.108 (0.033)	0.035 (0.104)	0.105 (0.034)	0.329 (0.128)	0.333 (0.193)
E₂	School attainment			0.145 (0.007)	0.125 (0.025)	0.141 (0.007)	0.127 (0.026)	0.137 (0.007)	0.038 (0.036)	0.008 (0.048)
E₃	Skilled job tenure							0.018 (0.006)	0.138 (0.018)	0.149 (0.023)
	Age at interview in 2002–2004	0.243 (0.090)	0.089 (0.104)	0.121 (0.102)	0.122 (0.085)	0.173 (0.097)	0.145 (0.086)	0.146 (0.097)	0.054 (0.099)	0.090 (0.119)
	Age at interview squared × 10	−0.041 (0.013)	−0.019 (0.015)	−0.021 (0.015)	−0.021 (0.013)	−0.029 (0.015)	−0.025 (0.013)	−0.025 (0.015)	−0.015 (0.015)	−0.021 (0.017)
	Constant	−3.032 (1.462)	−1.089 (1.597)	−2.348 (1.674)	−2.260 (1.419)	−2.956 (1.597)	−2.587 (1.397)	−2.513 (1.589)	−0.033 (1.590)	−0.451 (1.867)
	Test diagnostics
	F Statistic	25.0	18.6	166.1	23.1	118.8	17.1	96.5	31.1	22.7
	[p value]	[<0.01]	[<0.01]	[<0.01]	[<0.01]	[<0.01]	[<0.01]	[<0.01]	[<0.01]	[<0.01]
	Hansen J test [p value]		[<0.01]		[<0.01]		[<0.01]		[0.12]	[0.12]
	Hausman test [p value]		[0.10]		[0.87]		[0.98]		[<0.01]
	Panel C: First-stage diagnostics
	Kleibergen–Paap Wald F statistic		7.28		15.41		6.35		6.31
	[p value]		[<0.05]		[<0.05]		[<0.05]		[<0.05]

OLS, two-step instrumental variables GMM (IV-GMM), and limited information maximum likelihood (LIML) estimation using instruments in Table 2. Standard errors calculated allowing for clustering at the birth-year-community cohort level (64 clusters) are shown in parentheses below coefficient estimates, which are in bold if significant at 5 % or lower; p values are in brackets

Back to article page