The Effects of a Structured Curriculum on Preschool Effectiveness: A Field Experiment

Mari Rege; Ingunn Størksen; Ingeborg F. Solli; Ariel Kalil; Megan M. McClelland; Dieuwer ten Braak; Ragnhild Lenes; Svanaug Lunde; Svanhild Breive; Martin Carlsen; Ingvald Erfjord; Per Sigurd Hundeland

doi:10.3368/jhr.0220-10749R3

Article Figures & Data

Figures

Tables
Additional Files

Download figure
Open in new tab
Download powerpoint
Figure 1
Experimental Design
Notes: 71 preschool centers randomly split between control and treatment. Preschool year 2015–2016, teachers in treated centers attended the teacher training and helped revise the curriculum. In 2016–2017, teachers in treated centers implemented the structured curriculum with the five-year-olds in their center. In 2017–2018, teachers in control centers received the teacher training. We assessed children’s skills in language, mathematics, and executive functioning in August 2016 (baseline, T1), June 2017 (post-intervention, T2), and March 2018 (follow-up, T3).

View popup

Table 1

Test Score Correlations

	Summary Score	Mathematics	Executive Functioning	Language
T1 Test Scores
Mathematics	0.807
Executive functioning	0.837	0.545
Language	0.780	0.411	0.483
T2 Test Scores
Summary score	0.764	0.610	0.648	0.595
Mathematics	0.620	0.593	0.507	0.404
Executive functioning	0.667	0.521	0.667	0.428
Language	0.592	0.384	0.418	0.634
T3 Test Scores
Summary score	0.708	0.571	0.604	0.540
Mathematics	0.551	0.502	0.470	0.362
Executive functioning	0.595	0.483	0.586	0.374
Language	0.574	0.402	0.412	0.578

Notes: Gross sample, N ≤ 691. Sample varies slightly across test scores and across assessment period. All correlations are significant at the 1 percent level.

View popup

Table 2

Descriptive Statistics and Balance Test for T2 and T3 Sample

	Post-intervention T2					Follow-up T3
	Control	Treat	Difference	Wild P	N	Control	Treat	Difference	Wild P	N
T1 summary score	0.022	−0.021	−0.056	0.501	650	0.028	−0.016	−0.064	0.454	661
	(0.058)	(0.053)	(0.083)			(0.059)	(0.052)	(0.083)
T1 executive functioning	−0.015	0.009	0.024	0.799	650	−0.022	0.019	0.040	0.669	661
	(0.059)	(0.053)	(0.090)			(0.058)	(0.052)	(0.093)
T1 language	0.058	−0.046	−0.112	0.190	650	0.076	−0.050	−0.147^*	0.072	661
	(0.059)	(0.052)	(0.084)			(0.060)	(0.050)	(0.076)
T1 mathematics	0.010	−0.013	−0.046	0.573	650	0.014	−0.008	−0.047	0.578	661
	(0.056)	(0.055)	(0.079)			(0.055)	(0.053)	(0.084)
Female	0.517	0.480	−0.023	0.469	650	0.515	0.475	−0.029	0.435	661
	(0.500)	(0.500)	(0.032)			(0.500)	(0.500)	(0.037)
Birth month	6.878	6.807	−0.076	0.748	650	6.744	6.826	0.063	0.813	661
	(3.184)	(3.213)	(0.235)			(3.229)	(3.177)	(0.266)
Mother education	14.377	14.161	−0.169	0.464	626	14.385	14.126	−0.230	0.268	637
	(2.590)	(2.587)	(0.230)			(2.547)	(2.569)	(0.207)
Father education	13.814	13.696	−0.136	0.658	620	13.782	13.715	−0.085	0.776	628
	(2.563)	(2.504)	(0.265)			(2.536)	(2.469)	(0.252)
Mother earnings	341,408	320,168	−22,902	0.324	648	339,897	322,267	−21,206	0.329	658
	(221,464)	(206,324)	(22,751)			(214,889)	(203,739)	(21,584)
Father earnings	544,773	558,596	14,311	0.512	636	547,552	562,885	18,007	0.415	643
	(259,182)	(267,057)	(21,544)			(262,667)	(272,631)	(21,599)
Non-Western immigrant	0.130	0.201	0.068	0.123	650	0.136	0.203	0.064	0.142	661
	(0.337)	(0.401)	(0.040)			(0.343)	(0.403)	(0.039)
Missing T1 test scores	0.051	0.034	0.015	0.437	650	0.058	0.041	0.016	0.518	661
	(0.013)	(0.010)	(0.019)			(0.014)	(0.010)	(0.024)
N	292	358	650			293	368	661
F-test			0.295					0.000

Notes: ^*p < 0.1. The columns provide mean (standard deviation) for covariates and T1 test scores for the control group and treatment group in the T2 and T3 analytic samples. The column labeled “Difference” is the estimated coefficient (standard error) from regressing each covariate against treatment status. Regressions are clustered on and control for randomization block.

View popup

Table 3

Main Results—Treatment Effect on Test Scores at Post-Intervention (T2) and in the One-Year Follow-up (T3)

	Post-intervention (T2)				Follow-up (T3)
	Summary Score	Mathematics	Executive Functioning	Language	Summary Score	Mathematics	Executive Functioning	Language
Model 1
Treat	0.121^*	0.155	0.123^*	0.011	0.134^*	0.228^***	0.058	0.043
	(0.063)	(0.093)	(0.058)	(0.068)	(0.072)	(0.063)	(0.055)	(0.093)
Wild P	0.079	0.145	0.057	0.870	0.085	0.004	0.296	0.680
N	652	650	652	648	661	661	660	659
Adj. R²	0.612	0.440	0.492	0.532	0.522	0.364	0.382	0.492
Model 2
Treat	0.130	0.154	0.146^**	0.012	0.133	0.213^**	0.076	0.033
	(0.074)	(0.105)	(0.066)	(0.075)	(0.093)	(0.088)	(0.075)	(0.098)
Wild P	0.122	0.208	0.056	0.867	0.188	0.043	0.348	0.744
N	652	650	652	648	661	661	660	659
Adj. R²	0.147	0.114	0.114	0.101	0.123	0.083	0.084	0.176
Model 3
Treat	0.122^*	0.187^*	0.109	−0.001	0.118	0.236^***	0.063	−0.011
	(0.068)	(0.094)	(0.064)	(0.064)	(0.067)	(0.063)	(0.047)	(0.087)
Wild P	0.095	0.079	0.115	0.987	0.108	0.004	0.186	0.912
N	652	650	652	648	661	661	660	659
Adj. R²	0.613	0.433	0.492	0.521	0.517	0.349	0.379	0.456
Model 4
Treat	0.091	0.166	0.106	−0.055	0.091	0.213^**	0.064	−0.057
	(0.101)	(0.108)	(0.090)	(0.098)	(0.102)	(0.089)	(0.074)	(0.117)
Wild P	0.387	0.165	0.263	0.588	0.377	0.035	0.400	0.632
N	652	650	652	648	661	661	660	659
Adj. R²	0.017	0.004	0.022	0.015	0.015	0.026	0.007	0.031

Notes: ^*p < 0.1, ^**p < 0.05, ^***p < 0.01. Each column in each panel presents regression coefficient of treated (standard error) using ordinary least squares. For both assessment period, Model 1 regresses outcome on the treatment indicator, controlling for baseline test scores, gender, birth month, parental characteristics (mother and father’s education level, earnings, an indicator for non-Western country of birth), and indicators for late consent and not having participated in the T1 assessment. In Model 2, we exclude baseline test scores from the model. In Model 3, we restrict controls to baseline test scores and an indicator for not having participated in the T1 assessment. Model 4 has no controls. All regressions are clustered on and control for randomization block. We have utilized the boottest package in Stata to do a few-cluster-correction of the p-value, reported in the table as Wild P.

View popup

Table 4

Heterogeneous Treatment Effect on Test Scores at Post-intervention (T2) and in the One-Year Follow-up (T3) across High/Low Preschool Quality, Baseline Skills, and Parent Education

	Post-intervention (T2)				Follow-up (T3)
	Summary Score	Mathematics	Executive Functioning	Language	Summary Score	Mathematics	Executive Functioning	Language
Panel A: Preschool Center Quality
Treat	0.156	0.227	0.127	0.026	0.219^***	0.311^***	0.064	0.157^**
	(0.097)	(0.138)	(0.105)	(0.096)	(0.072)	(0.096)	(0.064)	(0.072)
Treat × High	−0.065	−0.113	−0.019	−0.038	−0.186^*	−0.153	−0.030	−0.264
	(0.107)	(0.164)	(0.126)	(0.115)	(0.102)	(0.124)	(0.116)	(0.152)
Wild P (treat)	0.142	0.170	0.261	0.791	0.027	0.025	0.330	0.034
Wild P (T × H)	0.545	0.522	0.883	0.736	0.102	0.246	0.796	0.132
N	638	636	638	634	648	648	647	646
Adj. R²	0.605	0.433	0.490	0.526	0.519	0.357	0.385	0.489
Mean low	−0.171	−0.152	−0.157	−0.105	−0.123	−0.098	−0.126	−0.078
Mean high	0.174	0.152	0.161	0.108	0.124	0.084	0.139	0.081
Panel B: Baseline Skills
Treat	0.066	0.121	0.104	−0.059	0.123	0.205^*	0.008	0.010
	(0.093)	(0.125)	(0.089)	(0.084)	(0.111)	(0.109)	(0.093)	(0.089)
Treat × High	0.107	0.049	0.026	0.148	0.023	0.035	0.088	0.068
	(0.123)	(0.143)	(0.100)	(0.137)	(0.108)	(0.133)	(0.133)	(0.135)
Wild P (treat)	0.503	0.401	0.261	0.481	0.282	0.088	0.933	0.908
Wild P (T × H)	0.395	0.741	0.795	0.288	0.821	0.784	0.511	0.608
N	652	650	652	648	661	661	660	659
Adj. R²	0.616	0.444	0.492	0.534	0.527	0.366	0.383	0.491
Mean low	−0.637	−0.496	−0.553	−0.547	−0.584	−0.442	−0.477	−0.469
Mean high	0.625	0.487	0.543	0.550	0.586	0.441	0.479	0.465
Panel C: Parent Education
Treat	0.086	0.141	0.032	0.040	0.114	0.250^**	−0.014	0.038
	(0.102)	(0.115)	(0.098)	(0.105)	(0.127)	(0.105)	(0.108)	(0.137)
Treat × High	0.060	0.010	0.163	−0.044	0.039	−0.042	0.108	0.036
	(0.121)	(0.128)	(0.119)	(0.100)	(0.136)	(0.131)	(0.157)	(0.123)
Wild P (treat)	0.400	0.240	0.759	0.692	0.394	0.038	0.902	0.799
Wild P (T × H)	0.618	0.938	0.206	0.647	0.787	0.756	0.506	0.791
N	641	639	641	637	649	649	648	647
Adj. R²	0.601	0.432	0.486	0.519	0.506	0.358	0.372	0.475
Mean low	−0.200	−0.192	−0.179	−0.117	−0.192	−0.174	−0.120	−0.176
Mean high	0.222	0.203	0.195	0.147	0.227	0.189	0.151	0.215

Notes: ^*p < 0.1, ^**p < 0.05, ^***p < 0.01. Each column in each panel presents regression coefficients with standard errors in parentheses, using ordinary least squares. The model specification is in line with Model 1 in Table 3. We add controls for gender, birth month and parental characteristics (education, earnings, and indicator for non-Western country of birth), baseline test scores, indicators for late consent and not having participated in the T1 assessment, and randomization block, with all regressions clustered on randomization block. Preschool center quality is measured as the preschool center fixed effect (center average covariate-adjusted test score). High/low center quality is split at median value. Parental education is measured as the average of mother’s and father’s number of years of education. 12 children with no information on parental education (balanced across treatment status) are excluded from the sample. High/low parental education and high/low baseline skills are split at median value. Mean low and mean high are means of outcome variable in relevant subsamples. We have utilized the boottest package in Stata to do a few-cluster-correction of the p-value, reported in the table as Wild P (treat) and Wild P (T×H).