Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Sep 1.
Published in final edited form as: Monogr Soc Res Child Dev. 2016 Jun;81(2):64–74. doi: 10.1111/mono.12239

IV. TESTING FOR DOSAGE-OUTCOME ASSOCIATIONS IN EARLY CARE AND EDUCATION

Yange Xue, Elizabeth B Miller, Anamarie Auger, Yi Pan, Margaret Burchinal, Hsiao-Chuan Tien, Ellen Peisner-Feinberg, Martha Zaslow, Louisa Tarullo
PMCID: PMC7462421  NIHMSID: NIHMS1617826  PMID: 27273507

Abstract

In this chapter, we turn to the question of whether there is evidence of an association between children’s development and the quantity or dosage of ECE across several large studies. As follow-up to the results summarized in the literature review, it is important to control adequately for selection effects in studying effects of dosage. There is also a need to examine different measures of dosage to see if consistent patterns of findings emerge across different measurement approaches. Accordingly, in this chapter, we will summarize analyses by using more rigorous approaches to controlling for selection than those used in previous research and will adopt several approaches to operationalizing dosage. Again, we are seeking replicated findings, as indicated in this section by similar significant findings across projects in analyses of dosage.

OVERVIEW OF ANALYTIC APPROACH IN DOSAGE ANALYSES

The first set of dosage analyses examines whether dosage measured as 2 years as opposed to 1 year of participation in Head Start is related to residualized gains in child outcomes based on FACES 2006 and 2009 and HSIS data. Propensity score matching was conducted to provide a more rigorous approach to controlling for selection by accounting for preexisting differences in measured variables (Shadish, Cook, & Campbell, 2002). Logistic regressions predicted whether children in the 3-year-old cohort stayed in Head Start for 1 or 2 years based on a set of family characteristics and children’s skills at Head Start entry across domains. The children in the 3-year-old cohort who stayed for 2 years were matched with children in the 4-year-old cohort who, by definition, had only 1 year of Head Start, using nearest-neighbor matching with replacement, based on predicted scores from the logistic regressions. The analyses of the matched children compared children with 1 or 2 years of Head Start and included the corresponding entry score and other child and family characteristics as covariates.

Representing different approaches to measuring dosage, the next set of dosage analyses considered attendance/absence in ECE, total number of hours per week in ECE, and observed time spent on instruction. Such analyses of dosage are based on regression models that extended the final models from the quality threshold analyses. In the analyses, the dosage variable was added to the final model from the quality threshold analyses for each quality measure. The model then included that quality measure as a piecewise predictor if indicated in the threshold analyses and, if not so indicated, as a linear predictor.

The final set of dosage analyses examined whether more time in higher quality care was related to larger gains in child outcomes. Using data from FACES 2009 and HSIS, we asked whether children who experienced 2 years of higher quality center-based care showed larger gains in child outcomes than other children with 2 years of center-based care. The same propensity score analysis strategy described above was employed to compare child outcomes of children with 2 years of high-quality center-based care versus other children with 2 years of center-based care. Using propensity score matching, we accounted for differences between the groups of children in terms of skills at entry to center-based care and family characteristics. The CLASS was used to determine quality of care in FACES 2009 and ECERS-R to determine quality of care in HSIS. Finally, we examined interactions between quality of care and three further measures of dosage—absences, hours per week of care, and time spent on instruction in specific content areas. Interactions between quality and dosage were tested for the various measures of quality.

RESULTS OF DOSAGE ANALYSES

One or Two Years of Head Start

Propensity score analyses were first conducted to examine the relationship between the number of years of exposure to Head Start and gains in child outcomes (Shadish, Cook, & Campbell, 2002). The propensity score analysis identified children who entered Head Start at age 4, experienced 1 year of Head Start, and had similar family characteristics and standardized entry skill levels as children who entered at age 3 and experienced 2 years of Head Start. Analyses compared the matched groups on child outcomes to test whether children showed higher skill levels at Head Start exit and in spring of kindergarten if they experienced 2 years instead of 1 year of Head Start. We conducted the analyses separately with two cohorts of FACES (FACES 2006 and 2009) and HSIS, allowing us to examine the extent to which findings replicated across cohorts and Head Start samples.

To reduce selection bias, the propensity score analyses balanced confounding covariates in the two groups of interest (in this analysis, the groups participating in Head Start for 1 versus 2 years). The analysis focused on identifying the characteristics of the group of 3-year-old children who participated in Head Start for 2 years and identifying children in the 4-year-old group with similar characteristics who participated in Head Start for a single year.

For analyses with FACES, the following variables were analyzed by using logistic regression to predict which children in the 3-year-old cohort remained in Head Start for 2 years: child race/ethnicity, gender, disability, household language, family poverty ratio, maternal education, employment, and depressive symptoms, household mobility, neighborhood safety, and child pretest standard scores on PPVT-4 and WJ III Letter-Word Identification, Spelling, and Applied Problems. The coefficients from the analyses were then used to create propensity scores for the 3-year-old cohort with 2 years of Head Start and the 4-year-old cohort with 1 year of Head Start. We used multiple imputations to account for missing data and used appropriate weights for the sampling design (Appendix A).

For conducting analyses with HSIS and estimating the propensity score (in other words, the conditional probability of enrollment in Head Start for 2 years), we included the following covariates that might be related to both enrollment in Head Start for 2 years and child outcomes: child race/ethnicity; gender; disability; household language; maternal education and depressive symptoms; whether the mother was a recent immigrant to the United States; three family structure variables, including whether both biological parents lived with the child, whether the child’s mother was married, and whether the mother was a teenager at the child’s birth; and child pretest standard scores on PPVT-3 and WJ III Letter-Word Identification, Spelling, and Applied Problems. The propensity score analysis involved equating on these covariates the children in HSIS who entered Head Start at age 3 and had 2 years of Head Start with 4-year-old children who had 1 year of Head Start (Appendix A).

Nearest-Neighbor Matching With Replacement

We used nearest-neighbor matching with replacement to match the 2-year group in the 3-year-old cohort with the 1-year group in the 4-year-old cohort (Shadish et al., 2002). For each child in the 2-year group, the potential comparison child in the one-year group with the closest absolute propensity score, or the “nearest neighbor,” was selected. Matching with replacement allows some children in the 1-year group to be used more than once to match to children in the 2-year group. In FACES 2006, the resulting sample after propensity score matching included 809–854 children in the 2-year group and 377–404 children in the 1-year group, with the total sample ranging from 1,084 to 1,118 (the sample size varies because of multiple imputation). Approximately half of the 1-year group children was matched to children in the 2-year group. In FACES 2009, the resulting sample after matching included 736–795 children in the 2-year group and 391–433 children in the 1-year group, with the total sample ranging from 1,209 to 1,246. More than half of the 1-year group was matched to children in the 2-year group. In HSIS, the resulting sample after matching included 714 children in the 2-year group and 770 children in the 1-year group, with a total sample of 1,484. Close to 85% of the 1-year group were matched to children in the 2-year group. In Table 1a and 1b of Appendix A, we show the descriptive statistics for those enrolled in Head Start for 1 year versus 2 years before and after matching. In all three samples, the children with 1 and 2 years of Head Start in the propensity matched samples did not differ significantly, and none of standardized mean difference was |.10| or greater on the child/family characteristics or baseline scores used in propensity score matching.

Next, multilevel analyses compared the matched children with 1 or 2 years of Head Start on each outcome. The multilevel analyses accounted for nesting of children in centers and included the same set of child and family covariates (Table 11). In both FACES 2006 and 2009, children with 2 years of Head Start scored significantly higher than children with 1 year on vocabulary skills at exit from Head Start and a year later at the end of kindergarten (.10 ≤ d ≤ .17). Two years compared to 1 year of Head Start were related to higher mathematics skills in FACES 2009 but not in FACES 2006, whereas 2 years compared to 1 year of Head Start were related to higher literacy skills in FACES 2006 but not in FACES 2009. In HSIS, children with 2 years of Head Start scored significantly higher than children with 1 year on literacy skills at exit from Head Start and at the end of kindergarten (.14 ≤ d ≤ .16). No evidence emerged suggesting that more years of Head Start were related to social skills or behavior problems.

TABLE 11.

Dosage: Comparing Children With 1 or 2 Years of Head Start in Propensity-Score Matched Samples: FACES 2006, FACES 2009, and HSIS

FACES 2006
 FACES 2009
 HSIS
Head Start Exit
 Spring Kindergarten
 Head Start Exit
 Spring Kindergarten
 Head Start Exit
 Spring Kindergarten
B (SE) d B (SE) d B (SE) d B (SE) d B (SE) d B (SE) d
Language: PPVTa 3.61*** (0.96) .16 2.77** (0.85) .15 3.49*** (.85) .17 1.88* (0.92) .10 0.07 (0.55) .00 1.15+ (0.63) .08
Literacy: WJ Letter Worda 6.00* (1.87) .14 2.74 (1.96) .08 3.61* (1.48) .11 3.94* (1.79) .10 2.36** (0.85) .16 2.06* (1.05) .14
Math: WJ Applied Problemsa 1.49 (1.25) .06 1.64 (1.53) .07 0.30 (0.48) .08 −0.02 (0.49) −.00 0.28 (0.76) .02 1.04 (1.02) .07
Social skillsb 0.44 (0.31) .08 0.03 (0.37) .01 −0.22 (0.36) −.03 −0.34 (0.47) −.04
Behavior problemsb −0.37 (0.38) −.06 −0.21 (0.59) −.02 −0.22 (0.36) −.03 −0.34 (0.47) −.04 −0.22 (0.65) −.01 −0.39 (0.61) −.03
Open in a new tab

Note. FACES 2006 = Family and Child Experiences Survey-2006; FACES 2009 = Family and Child Experiences Survey-2009; HSIS = Head Start Impact Study; WJ = Woodcock Johnson.

+

.05 < p < .1

*

p < .05

**

p < .01

***

p < .001.

Estimates are weighted. Covariates include corresponding pretest score, child age, gender, race/ethnicity, household language, poverty ratio, maternal education, employment, and depressive symptoms, household mobility, and neighborhood safety. Missing data were handled using multiple imputation (N = 10). Children in Head Start for 2 years were matched with those in Head Start for one year based on propensity scores estimated from baseline characteristics and scores (nearest neighbor matching with replacement).

a

Controlled for baseline (fall of first year) standard scores on corresponding measures.

b

Did not control for baseline scores.

Sensitivity analyses were conducted. In addition to nearest-neighbor matching, we conducted robustness testing by using two other propensity score approaches: caliper matching and propensity score weighting (Shadish et al., 2002). In our case, caliper matching selected all 4-year-old children whose characteristics are sufficiently close in propensity score units to those of a 3-year-old child with 2 years of Head Start. We carried out caliper matching with replacement so that a potential comparison child in the 1-year group may be matched to several children in the 2-year group. The findings are similar to the results from nearest-neighbor matching (not reported).

We tried to weight the 2-year and 1-year groups by using the inverse probability of treatment (the inverse of the propensity score for the 2-year group and the inverse of one minus the propensity score for the 1-year group). Again, the findings are similar to the results from nearest-neighbor matching (not reported).

In summary, the analyses provide replicated evidence that 2 years of Head Start appear to have a larger impact on children’s academic but not social skills than a single year, both at program exit and 1 year later in spring of kindergarten.

Attendance/Absence

The next set of analyses examined whether absences predicted residualized gains in child outcomes. Absences were added to the final model from the quality threshold analyses. Again, separate analyses were conducted for each quality measure, using that quality indicator as a piecewise predictor if indicated in the threshold analyses and otherwise as a linear predictor. The analyses included the same set of covariates, including the child’s fall score on the outcome.

Two data sets—FACES 2006 and NC-PK—measured absences. FACES 2006 measured the number of days the child was absent based on teacher and parent reports, whereas NC-PK recorded attendance based on a daily teacher report. The data from both projects were recoded into the FACES categories (1 = never; 2 = 1–5 days; 3 = 6–10 days; 4 = 11–20 days; 5 = more than 20 days). The first set of columns in Table 12 gives the results, listing the smallest coefficient for absences from the analyses of that outcome that used different quality measures. We focused on the smallest coefficient for absences from the separate models with different measures of quality because that model provided the most conservative test based on the assumption that the quality variable in that model accounted for the greatest variance in outcomes (all results are available on request). As shown in Table 12, children with more absences according to attendance reports had smaller residualized gains in language in NC-PK, in mathematics according to both teacher and parent absence reports in FACES, and in literacy according to teacher absence reports in FACES. Thus, the findings provide some evidence that children with more absences show lower levels of academic but not social skills, although the specific academic outcomes related to absences varied across studies and informants.

TABLE 12.

Dosage: Associations Between Child Outcomes and Absences, Hours/Week of Care, and Instruction Time

Absences
 Hours/Week
 Time in Reading Instruction
 Time in Math Instruction
FACES 2006-Teacher Report FACES 2006-Parent Report NC-PK-Class Records FACES 2006 NCEDL NCEDL PCER NCEDL PCER
Language: PPVT
 B (SE) −.44+ (.25) −.27 (.28) −.35* (.21) −.00 (.03) −0.01 (0.02)
 d −.03 −.01 −.07 −.00 −.01
Math WJ-AP
 B (SE) −1.16*** (.32) −.89* (.35) −.53+ (.30) −.00 (.04) −0.01 (0.02) 8.21* (3.26) 1.35** (0.51)
 d −.07 −.05 −.04 −.00 −.01 .04 .07
Literacy: WJ-LW
 B (SE) −1.01** (.35) −.38 (.39) −.38 (.48) 09+ (.05) −0.02 (0.02) 7.48* (2.81) .80*** (.17)
 d −.06 −.02 −.03 .05 −.01 .04 .11
Social skills
 B (SE) −.11 (.08) −.01 (.08) −.14 (.33) −.02+ (.01) −0.02 (0.01)
 d −.02 −.00 −.01 −.04 −.03
Behavior problems
 B (SE) .23* (.10) .067 (.10) −.64+ (.34) .03+ (.02) 0.01 (0.01)
 d .04 .01 −.05 .04 .02
Open in a new tab

Note. FACES 2006 = Family and Child Experiences Survey-2006; NC-PK = More-at-Four NC Pre-kindergarten Evaluation; NCEDL = NCEDL 11 State Pre-kindergarten Study; PCER = Preschool Curriculum Evaluation Research; PPVT = Peabody Picture Vocabulary Test; WJ-AP = Woodcock Johnson Letter-Applied Problems; MJ-LW = Woodcock Johnson Letter-Word Identification.

+

p < .10

*

p < .05

**

p < .01

***

p < .001.

The coefficient reported in the table is the coefficient for absences in the model in which the quality variable accounted for the most variance. Separate analyses examined each quality measure as a covariates—using the best model from the threshold analysis for that study using that outcome and quality measure. In no case was a different inference draw in the analyses that involved the other quality variables. The other covariates include fall score on same outcome, gender, race, time between fall and spring assessments, whether child speaks English at home, and if relevant site, whether program was a Head Start program and whether the program was located in a school. Quality measures included ECERS-R Total and Interaction in FACES, NC-PK, NCEDL, and PCER; CLASS Instructional Support in FACES 2006 & 2009, NC-PK, and NCEDL, CLASS Emotional Support in NC-PK & NCEDL, and TBRS in PCER. Coefficients reported analyses involving all classroom quality analyses—reporting the largest and smallest coefficients from those analyses. Control variables include gender, race, maternal education, whether or not below poverty line, Head Start, whether program was located in a public school, state and fall assessment.

Number of Hours Per Week

The next set of analyses used the same strategy to examine associations between hours per week and child outcomes, using data from the two studies in which hours per week were reported. The program director or classroom teacher reported hours of operation in FACES 2006 and NCEDL; hours varied across programs in both studies. We added hours per week of ECE to the final quality threshold model, with results of analyses reported in the middle set of columns in Table 12. Again, separate analyses of each outcome were conducted by using each quality measure, adding hours per week to the final model in the threshold analyses involving HLMs, covariates, and multiple imputations. For a given outcome, we report the coefficient in which quality accounted for the greatest variance from the models involving different quality measures. Hours per week of operation of ECE programs did not emerge as a statistically significant predictor; thus, no evidence of replicated associations emerged from the analyses.

Instructional Time Within Specific Content Areas

The same strategy was used to examine instructional time, drawing on data from two studies. For NCEDL, we used the proportion of time observed in instruction on Snapshot; for PCER, we used the rating of time observed in instruction according to TBRS. In both data sets, we examined instruction time in mathematical activities as a predictor of mathematics outcomes and instruction time in literacy activities as a predictor of language and literacy outcomes. Results are in the final set of columns in Table 12. The analyses also used the final model from the threshold analyses, including each quality measure as either a piecewise or linear predictor, the same covariates, HLM analyses, and multiple imputation. The analyses allow us to consider time on instruction in a particular content area, holding constant the observed measure of quality included in the threshold analysis.

As shown in the final columns of Table 12, time spent in mathematics instruction was a consistent predictor of mathematics skills (NCEDL: d = .04; PCER: d = .07), and time spent in literacy instruction was a consistent predictor of literacy outcomes (NCEDL: d = .04–.05; PCER: d = .11), even after accounting for quality of instruction, the child’s skills in the fall, and demographic covariates.

Quality by Dosage Interactions

The next set of analyses tested whether there were interactions between quality and dosage. Using FACES 2009 and HSIS data, we first used the propensity score approach to examine whether children with more years of high-quality care had better outcomes. In another set of analyses, we added the interaction between the final quality terms in the quality threshold analysis and hours per week, absence, and time spent in specific instructional domains.

Number of Years of High-Quality ECE

The propensity score analyses of the FACES 2009 and HSIS data examined 2 years of high-quality center-based care. The analyses examined whether children’s academic and social skills were higher when children experienced 2 years of high-quality care (as opposed to only one or no years of high-quality care) among children with 2 years of center-based ECE. In both studies, the analyses included the cohort of 3-year-old children with 2 years of ECE. The children were classified into two groups based on whether they experienced high-quality care in both years. Propensity score matching identified matches for children with 2 years of high-quality care on children’s entry skills and family characteristics (child age, gender, race, disability, household language, income-to-needs ratio, maternal education, maternal depression, maternal employment, single-parent family, household mobility, neighborhood safety, as well as language, literacy, mathematics, and social skills based on the fall 3-year-old assessments). Propensity score matching using nearest-neighbor matching eliminated initial differences between the groups on these variables. We attempted to use the same cut-points as in the threshold analyses but were not able to identify a sufficient number of children with high-quality care in FACES based on those criteria, especially on the Instructional Support scale. Instead, we used the criteria that Head Start uses for its monitoring system; that is, a low-quality classroom was defined with scores of lower than 2 on Instructional Support, 5 on Emotional Support, and 4 on Classroom Organization, with 258 children in the high-quality group and 419 in the comparison group after matching. In HSIS, a high-quality classroom was defined with a score above 4.5 on the ECERS-R Total score, with 361 children in the high-quality group and 177 in the comparison group for the group with fewer than 2 years of high-quality in HSIS.

Results in Table 13 suggest that the outcomes of children with 2 years of high-quality care in Head Start did not differ from those of other children at Head Start exit and spring of kindergarten.

TABLE 13.

Testing Dosage Threshold: Comparing FACES 2009 Children With 2 Years of High-Quality Care and <2 Years of High Quality Care Using Propensity Score Matchinga,b

FACES 2009a
 HSISb
Head Start Exit, B (SE) Spring Kindergarten, B (SE) Head Start Exit, B (SE) Spring Kindergarten, B (SE)
Language: PPVT 0.95 (1.51) 0.67 (1.60) 0.37 (0.85) 0.55 (0.066)
Literacy WJ-LW 0.41 (1.55) 1.53 (1.50) 0.02 (1.15) 0.80 (1.16)
Math WJ-AP 0.06 (1.55) −0.27 (1.56) −0.92 (1.04) 2.18+ (1.27)
Behavior problems −0.44 (0.46) 0.00 (0.78)
Social skills 0.76 (0.51) −0.08 (0.70)
Open in a new tab

Note. FACES 2009 = Family and Child Experiences Survey-2009; HSIS = Head Start Impact Study; PPVT = Peabody Picture Vocabulary Test; WJ-LW = Woodcock Johnson Letter-Word; WJ-AP = Woodcock Johnson Applied Problems.

+

.1 < p < .05.

Matched on: child age, gender, race, disability, household language, income to needs ratio, maternal education, maternal depression, maternal employment, single-parent family, household mobility, neighborhood safety, and baseline scores.

a

FACES: High-quality classroom is defined having scores of >2 on Instructional Support, 5 on Emotional Support, and 4 on Classroom Organization. Sample sizes range n = 250–258 for the group with 2 years of high quality and n = 410–419 for the group with less than 2 years of high quality.

b

HSIS: High-quality classroom is defined as having score above 4.5 on ECERS-R total score. There were n = 361 with 2 years of high quality and n = 177 for the group with less than 2 years of high quality in HSIS.

Interactions Between Quality and Dosage

We tested the interaction between the final quality terms in the threshold analysis and absences, hours per week, and time spent in content-specific instruction. Only one of the interactions was observed across studies and, therefore, meets our goal of obtaining replicated evidence.

First, we examined interactions between absences and quality. Significant interactions between quality and absences emerged in the analysis of FACES 2006, but not in the analyses of the other two studies with absence data. In higher quality care according to the ECERS-R Total score, teacher-reported absence in FACES 2006 was a stronger negative predictor of the residualized gains in language (PPVT-4) and mathematics (WJ III Applied Problems). No interactions were statistically significant in FACES 2009 or NC-PK.

No evidence of interactions emerged in analyses examining hours per week of ECE and quality based on data from NCEDL. Inconsistent findings emerged in analyses examining associations between hours per week of care and quality in relation to child outcomes based on FACES 2006 data. Findings suggested that hours per week was a stronger predictor of PPVT in higher quality care according to the ECERS total and a weaker predictor of ECLS-B mathematics skills in higher quality care according to CLASS instructional support. No evidence supporting interactions emerged in analyses of NCEDL data.

Finally, interactions between time spent in instruction and quality of care were examined. In both NCEDL and PCER, interactions between time spent in mathematics activities and quality of mathematics instruction were statistically significant. Results indicated that more time in mathematics instruction was a stronger predictor of residualized gains in mathematics skills (WJ Applied Problems) when quality of instruction was higher according to the CLASS instructional support in NCEDL and according to the TBRS rating of the quality of mathematics instruction in PCER. No evidence of interactions emerged in analyses of the quality and quantity of literacy instruction.

In summary, there was very limited evidence in the analyses of an interaction of quality and dosage. The only replicated finding pointed to greater gains in mathematics skills when time on mathematical instruction was higher, especially in programs with higher quality instruction.

SUMMARY OF RESULTS FOR DOSAGE ANALYSES

The dosage analyses provided evidence that dosage of ECE was related to the acquisition of academic skills. Propensity score analyses indicated that academic skills were enhanced with an additional year in Head Start, but not necessarily with an additional year in higher quality care among children with 2 years of center-based ECE. Children’s academic skills were also stronger when they were in classrooms where more time was spent in instruction. Analyses of absences provided some replicated evidence that children with fewer absences showed larger gains across outcomes in one study, but not across studies. Hours per week of the center-based programs were not reliably or consistently related to child outcomes. Finally, replicated findings of interactions between quality and dosage emerged only in analyses of mathematics outcomes as a function of the quality and quantity of mathematics instruction. Associations between dosage and child outcomes tended to be modest in all of these analyses.

APPENDIX

This article is part of the issue “Quality Thresholds, Features, and Dosage in Early Care and Education: Secondary Data Analyses of Child Outcomes,” Burchinal, Zaslow, and Tarullo (Issue Editors). For a full listing of articles in this issue, see: http://onlinelibrary.wiley.com/doi/10.1111/mono.v81.2/issuetoc.

TABLE 1A.

Comparing Child and Family Characteristics in Unmatched and Matched FACES Samples

FACES 2006
 FACES 2009
Before Matching
 After Matching
 Not Matched
 Before Matching
 After Matching
 Not Matched
2 Years in Head Start 1 Year in Head Start t-Test 2 Years in Head Start 1 Year in Head Start t-Test 1 Year in Head Start 2 Years in Head Start 1 Year in Head Start t-Test 2 Years in Head Start 1 Year in Head Start t-Test 1 Year in Head Start
Gender (Boy) % 51.38 53.46 51.05 56.32 51.62 51.38 47.96 51.31 47.78 47.64
Race/Ethnicity
 European-Am % 22.17 27.32 * 22.42 24.79 26.93 22.77 22.77 22.8 20.15 24.43
 Black % 43.54 28.44 *** 42.83 46.6 20.32 43.87 35.22 *** 43.79 42.22 30
 Hispanic % 25.19 35 *** 25.62 20.14 43.42 24.55 34.61 *** 24.59 28.91 38.82
 Asian % 1.04 1.71 1.06 0.81 2.14 1.76 1.86 1.76 2.25 1.67
 Other % 7.99 7.53 8.01 8.14 6.5 7.05 5.45 7.06 6.06 4.99
Child with a disability diagnosis % 5.74 4.75 5.7 5.61 4.20 3.24 2.62 3.24 3.66 2.17
Household language (not English) 13.71 28.02 *** 13.93 12.41 37.91 14.82 22.62 *** 14.84 17.31 26.62
Poverty ratio 2.74 2.78 2.74 2.71 2.84 2.58 2.56 2.57 2.56 2.55
Maternal education
 <High school % 30.78 35.7 * 30.78 30.03 39.47 25.64 33.74 *** 25.68 27.11 38.07
 High school/GED % 32.76 34.22 32.41 37.5 30.97 39.77 36.92 39.81 40.7 34.9
 Some college % 29.96 24.54 * 30.42 27.94 23.76 28.34 23.98 28.27 25.58 22.66
 BA+ % 6.82 6.28 6.73 5.89 6.33 6.59 5.58 6.58 6.72 4.46
Maternal employment % 56.89 50.86 * 56.75 57.35 47.31 52.8 52.02 52.81 50.95 52.15
Maternal depressive symptoms 5.76 5.09 * 5.67 6.08 4.49 5.05 4.93 5.06 5 4.89
Single parent % 48.62 48.33 48.89 51.22 46.85 54.2 50.71 54.23 50.18 51.42
Household mobility % 22.85 23.44 22.72 19.37 24.74 40.53 50.32** 40.59 37 60.04
Neighborhood safety M 12.9 12.52 12.52 14.04 12.1 10.48 10.78 10.49 9.49 11.67
Pretest scores
 PPVT standard score M 85.55 82.87 *** 85.54 85.47 81.08 86.69 85.52 86.7 85.49 85.48
 WJ letter word identification M 93.55 92.22 93.52 94.28 91.31 96.64 94.21** 96.57 96 93.09
 WJ spelling M 98.35 90.45 *** 97.82 96.5 87.54 95.18 94.52 95.15 94.51 94.45
 WJ applied problems M 92.2 83.99 *** 91.69 90.63 80.55 87.26 87.13 87.26 86.09 87.71
Sample size 822–868 759–810 809–854 377–404 383–406 737–796 741–809 736–795 391–433 349–376
Open in a new tab

Note. FACES 2006 = Family and Child Experiences Survey-2006; FACES 2009 = Family and Child Experiences Survey-2009; PPVT = Peabody Picture Vocabulary Test; WJ = Woodcock Johnson. Children enrolled in Head Start for two years were matched to those enrolled in Head Start for one year based on propensity scores using nearest neighbor matching with replacement.

*

p < .05.

**

p < .01.

***

p < .001.

TABLE 1B.

Descriptive Statistics For the Matched and Unmatched Child Samples: HSIS

HSIS
Before Matching
 After Matching
 Not Matched
2 Years in Head Start 1 Year in Head Start t-Test 2 Years in Head Start 1 Year in Head Start t-Test 1 Year in Head Start
Gender (Boy) % 48.57 50.97 48.88 50.84 48.49
Race/Ethnicity
 European American % 31.16 35.14 29.69 29.55 33.48
 Black % 35.78 19.69 *** 36.28 36.56 32.92
 Hispanic % 33.06 45.17 *** 34.03 33.89 33.59
Child Disability % 14.56 13.64 14.57 13.17 10.94
Household language (not English) % 26.8 37.71 *** 24.93 24.09 26.28
Maternal education
 Less than high school % 33.61 42.21 *** 33.61 31.23 36.89
 High school/GED % 34.56 31.4 34.31 35.71 33.42
 Beyond high school % 31.84 26.38 * 32.07 33.05 29.69
Maternal depressive symptoms % 19.86 21.75 20.44 21.57 19.25
Mother teenaged at birth % 13.61 17.63 * 14.01 13.59 16.02*
Mother married % 44.08 45.69 42.72 43.42 45.65
Both biological parents in household % 50.34 52.77 49.16 49.3 49.55
Mother recent immigrant % 14.83 23.81 *** 14.15 14.43 17.13*
Urban location % 82.86 85.07 82.35 81.93 82.92
Pretest scores
 PPVT M 91.4 90.85 91.4 91.78 91.83
 WJ letter word ID M 91.52 92.28 91.52 91.94 91.2
 WJ applied problems M 90.11 90.02 90.11 89.67 89.69
Sample size N 807 918 777 735 141
Mother teenaged at birth 13.61 17.63* 14.01 13.59 16.02*
Mother married 44.08 45.69 42.72 43.42 45.65
Biological parents live together 50.34 52.77 49.16 49.3 49.55
Mother recent immigrant 14.83 23.81*** 14.15 14.43 17.13*
Urban location 82.86 85.07 82.35 81.93 82.92
Pretest scores
 PPVT standard score 91.4 90.85 91.4 91.78 91.83
 WJ letter word identification standard score 91.52 92.28 91.52 91.94 91.2
 WJ applied problems standard score 90.11 90.02 90.11 89.67 89.69
Sample size 807 918 777 735 141
Open in a new tab

Note. HSIS = Head Start Impact Study; PPVT = Peabody Picture Vocabulary Test; WJ = Woodcock Johnson. Children enrolled in Head Start for two years were matched to those enrolled in Head Start for one year based on propensity scores using nearest neighbor matching with replacement.

*

p < .05.

**

p < .01.

***

p < .001.

REFERENCES

  1. Achenbach TM (1991). Manual for the child behavior checklist/4–18 and 1991 profiles. Burlington: University of Vermont, Department of Psychiatry. [Google Scholar]
  2. Allard Agnamba LT (2012). Preparation and ongoing support for early childhood instructional coaches: A case study exploration of an instructional coaching program. Philadelphia: University of Pennsylvania. [Google Scholar]
  3. Belsky J, & Pluess M (2012). Differential susceptibility to long-term effects of quality of child care on externalizing behavior in adolescence? International Journal of Behavioral Development, 36 (1), 2–10. [Google Scholar]
  4. Berry DJ, Bridges LJ, & Zaslow MJ (2004). Early Childhood Measures Profiles. Washington, DC: Child Trends, http://aspe.hhs.gov/hsp/ecmeasures04/report.pdf [Google Scholar]
  5. Bierman BL, Domitrovich CE, Nix RL, Gest SD, Welsh JA, Greenberg MT, et al. (2008). Promoting academic and social-emotional school readiness: The Head Start REDI program. Child Development, 79, 1802–1817. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Borenstein M, Hedges L, Higgins J, & Rothstein H (2005). Comprehensive meta-analysis version 2. Biostat: Englewood, NJ. [Google Scholar]
  7. Bryant D (2010). Observational measures of quality in center-based early care and education programs (OPRE Research-to-Policy Brief #5). Washington, DC: U.S. Department of Health, Human Services, Administration for Children, Families, Office of Planning, Research, Evaluation, and Child Trends. [Google Scholar]
  8. BUILD Initiative and Child Trends. (2014, November 3). A Catalog and Comparison of Quality Rating and Improvement Systems (QRIS). Retrieved November 7, 2014, from http://qriscompendium.org/ [Google Scholar]
  9. Burchinal M, Kainz K, & Cai Y (2011). How well do our measures of quality predict child outcomes? A meta-analysis and coordinated analysis of data from large-scale studies of early childhood settings In Zaslow M(Ed.), Reasons to take stock and strengthen our measures of quality (pp. 11–31). Baltimore, MD: Brookes Publishing. [Google Scholar]
  10. Burchinal P, Kainz K, Cai K, Tout K, Zaslow M, Martinez-Beck I, & Rathgeb C (2009). Early care and education quality and child outcomes (OPRE Research-to-Policy Brief #1). Washington, DC: U.S. Department of Health and Human Services, Administration for Children and Families, Office of Planning, Research and Evaluation, and Child Trends. [Google Scholar]
  11. Burchinal M, Magnuson K, Powell D, & Hong SS (2014). Early child care and education and child development In Bornstein M, Lerner R, & Leventhal T (Eds.), Handbook of child psychology and developmental science (pp. 223–267). New York, NY: Wiley. [Google Scholar]
  12. Burchinal M, Vandergrift N, Pianta R, & Mashburn A (2010). Threshold analysis of association between child care quality and child outcomes for low-income children in prekindergarten programs. Early Childhood Research Quarterly, 25, 166–176. [Google Scholar]
  13. Burchinal M, Vernon-Feagans L, Vitiello V, Greenberg M, & the Family Life Project Key Investigators. (2014). Thresholds in the association between child care quality and child outcomes in rural preschool children. Early Childhood Research Quarterly, 29, 41–51. [Google Scholar]
  14. Bureau of Labor Statistics (2013). Employment Characteristics of Families-2012 (USDL-13–0730) Retrieved April 15, 2016, from http://www.bls.gov/news.release/pdf/famee.pdf
  15. Caldwell B, & Bradley R (1984). Home Observation for Measurement of the Environment (HOME) inventory-revised edition. Little Rock: University of Arkansas. [Google Scholar]
  16. Camilli G, Vargas S, Ryan S, & Barnett WS (2010). Meta-analysis of the effects of early education interventions on cognitive and social development. Teachers College Record, 112, 579–620. [Google Scholar]
  17. Campbell FA, Pungello E, Miller-Johnson S, Burchinal M, & Ramey CT (2001). The development of cognitive and academic abilities: Growth curves from an early childhood education experiment. Developmental Psychology, 37, 231–242. [DOI] [PubMed] [Google Scholar]
  18. Campbell FA, Ramey CT, Pungello E, Sparling J, & Miller-Johnson S (2002). Early childhood education: Young adult outcomes from the Abecedarian Project. Applied Developmental Science, 6, 42–57. [Google Scholar]
  19. Chang H (2008). Present, engaged and accounted for: The critical importance of addressing chronic absence in the early grades. New York, NY: National Center for Children in Poverty. [Google Scholar]
  20. Clements DH, & Sarama J (2008). Experimental evaluation of the effects of a research-based preschool mathematics curriculum. American Educational Research Journal, 45, 443–494. [Google Scholar]
  21. Cleveland WS, & Devlin SJ (1988). Locally weighted regression: An approach to regression analysis by local fitting. Journal of the American Statistical Association, 83(403), 596–610. [Google Scholar]
  22. Connolly F (2012). Early elementary performance and attendance in baltimore city school’s prekindergarten and kindergarten. Baltimore, MD: Baltimore Education Research Consortium. [Google Scholar]
  23. Dearing E, McCartney K, & Taylor B (2009). Does higher-quality early child care promote low-income children’s math and literacy achievement in middle childhood? Child Development, 80, 1329–1349. [DOI] [PubMed] [Google Scholar]
  24. Duncan GJ, Dowsett CJ, Claessens A, Magnuson K, Huston AC, Klebanov P, et al. (2007). School readiness and later achievement. Developmental Psychology, 43(6), 1428. [DOI] [PubMed] [Google Scholar]
  25. Duncan GJ, & Magnuson K (2013). Investing in preschool programs. The Journal of Economic Perspectives, 27(2), 109–132. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Dunn LM, & Dunn DM (2007). PPVT-4: Peabody picture vocabulary test. Minneapolis, MN: NCS Pearson. [Google Scholar]
  27. Dunn LM, Dunn DM, & Schlichting JEPT (2005). Peabody picture vocabulary test-III-NL. Circle Pines, MN: American Guidance Service. [Google Scholar]
  28. Ehrlich S (2013). Preschool attendance in Chicago public schools: Relationships with learning outcomes and reasons for absences. University of Chicago: Consortium on Chicago School Research. [Google Scholar]
  29. Elicker JG, Langill CC, Ruprecht KM, Lewsader J, & Anderson T (2011). Final report: Evaluation of paths to QUALITY, Indiana’s child care quality rating and improvement system. Purdue University: Department of Human Development and Family Studies. [Google Scholar]
  30. Fantuzzo JW, Gadsden VL, & McDermott PA (2011). An integrated curriculum to improve mathematics, language, and literacy for Head Start children. American Educational Research Journal, 48, 763–793. [Google Scholar]
  31. Farnworth M, Schweinhart LJ, & Berrueta-Clement JR (1985, Fall). Preschool intervention, school success and delinquency in a high-risk sample of youth. American Educational Research Journal, 22, 445–464. [Google Scholar]
  32. Farver J, Lonigan C, & Eppe S (2009). Effective early literacy skill development for young Spanish-speaking English language learners: An experimental study of two methods. Child Development, 80, 703–719. [DOI] [PubMed] [Google Scholar]
  33. Gordon RA (2014). Assuring quality preschool: Where are we and where do we need to go? Presented in the American Educational Research Association Presidential Session on Universal Preschool: What Have We Learned, and What Does It Mean for Practice and Policy? (Chair: Rachel A. Gordon, Discussant: Libby Doggett), Philadelphia, PA. [Google Scholar]
  34. Grehan A, & Smith L (2004). Early literacy observation tool. Memphis, TN: University of Memphis Center for Research in Educational Policy. [Google Scholar]
  35. Gresham FM, & Elliott SN (1990). Social skill rating system. Circle Pines, MN: American Guidance Services. [Google Scholar]
  36. Hamre B (2012). Enhancing teachers’ intentional use of effective interactions with children In Pianta RC (Ed.), Handbook of early childhood education (pp. 507–532). New York, NY: Guilford Press. [Google Scholar]
  37. Hamre BK, Pianta RC, Burchinal M, & Downer JT (2010, March). A course on supporting early language and literacy development through effective teacher-child interactions: Effects on teacher beliefs, knowledge and practice. Paper presented at the Annual Meeting of the Society for Research on Educational Effectiveness, Washington, DC. [Google Scholar]
  38. Harms T, Clifford RM, & Cryer D (1998). Early Childhood Environment Rating Scale. New York, NY: Teachers College Press. [Google Scholar]
  39. Harms T, Cryer D, & Clifford RM (2006). The Infant/Toddler Environment Rating Scale. New York, NY: Teachers College Press. [Google Scholar]
  40. Harms T, Cryer D, & Clifford RM (2007). The Child Care Environment Rating Scale. New York, NY: Teachers College Press. [Google Scholar]
  41. Hightower AD, Work WC, Cowen EL, Lotyczewski BS, Spinell AP, Guare JC, et al. (1986). The teacher-child rating scale: A brief objective measure of elementary children’s school problem behaviors and competencies. School Psychology Review, 15, 393–409. [Google Scholar]
  42. Hill JL, Brooks-Gunn J, & Waldfogel J (2003). Sustained effects of high participation in an early intervention for low-birth-weight premature infants. Developmental Psychology, 39(4), 730–744. [DOI] [PubMed] [Google Scholar]
  43. Howes C, Burchinal M, Pianta R, Bryant D, Early D, Clifford R, et al. (2008). Ready to learn? Children’s pre-academic achievement in pre-kindergarten programs. Early Childhood Research Quarterly, 23, 27–50. [Google Scholar]
  44. Howes C, Phillips DA, & Whitebook M (1992). Thresholds of quality: Implications for the social development of children in center-based child care. Child Development, 63, 449–460. [DOI] [PubMed] [Google Scholar]
  45. Hubbs-Tait L, Culp AM, Huey E, Culp R, Starost H, & Hare C (2002). Relation of Head Start attendance to children’s cognitive and social outcomes: Moderation by family risk. Early Childhood Research Quarterly, 17, 539–558. [Google Scholar]
  46. Kaiser A, Dickinson D, Roberts M, Darrow C, Freiberg J, & Hofer K (2011). The effects of two language-focused preschool curricula on children’s achievement through first grade. Washington, DC: Society for Research on Educational Effectiveness. [Google Scholar]
  47. Karoly LA, Kilburn MR, & Cannon JS (2005). Early childhood interventions: Proven results, future promise. Santa Monica, CA: RAND Corporation. [Google Scholar]
  48. Keys TD, Farkas G, Burchinal MR, Duncan GJ, Vandell DL, Li W, et al. (2013). Preschool center quality and school readiness: Quality effects and variation by demographic and child characteristics. Child Development, 84 (4), 1171–1190. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Landry SH, Anthony JL, Swank PR, & Monesque-Bailey P (2009). Effectiveness of comprehensive professional development for teachers of at-risk preschoolers. Journal of Educational Psychology, 101, 448–465. [Google Scholar]
  50. Landry SH, Crawford A, Gunnewig SB, & Swank PR (2002). Teacher behavior rating scale. Unpublished instrument, University of Texas Health Science Center at Houston, Center for Improving the Readiness of Children for Learning and Education. [Google Scholar]
  51. Laughlin L (2013). Who’s minding the kids? child care arrangements: Spring 2011. Washington, DC: U.S. Census Bureau. [Google Scholar]
  52. Loeb S, Fuller B, Kagan SL, & Carrol B (2004). Child care in poor communities: Early learning effects of type, quality, and stability. Child Development, 75, 47–65. [DOI] [PubMed] [Google Scholar]
  53. Love JM, Kisker EE, Ross C, Raikes H, Constantine J, Boller K, et al. (2005). The effectiveness of Early Head Start for 3-year-old children and their parents: Lessons for policy and programs. Developmental Psychology, 41(6), 885. [DOI] [PubMed] [Google Scholar]
  54. Mashburn AJ, Downer JT, Hamre BK, Justice LM, & Pianta RC (2010). Consultation for teachers and children’s language and literacy development during pre-kindergarten. Applied Developmental Science, 14, 179–196. [DOI] [PMC free article] [PubMed] [Google Scholar]
  55. Mashburn AJ, Pianta RC, Hamre BK, Downer J, Barbarin OA, Bryant D, et al. (2008). Measures of classroom quality in pre-kindergarten and children’s development of academic, language and social skills. Child Development, 79, 732–749. [DOI] [PubMed] [Google Scholar]
  56. McCartney K, Burchinal M, Clarke-Stewart A, Bub KL, Owen MT, Belsky J, et al. (2010). Testing a series of causal propositions relating time in child care to children’s externalizing behavior. Developmental Psychology, 46, 1–17. [DOI] [PubMed] [Google Scholar]
  57. Moiduddin E, Aikens N, Tarullo L, West J, & Xue Y (2012). Child outcomes and classroom quality in FACES 2009. (OPRE Report 2012–37a). Administration for Children & Families. Retrieved October 15, 2014, from http://www.acf.hhs.gov/sites/default/files/opre/faces_2009.pdf [Google Scholar]
  58. National Institute of Child Health and Human Development (NICHD) Early Child Care Research Network (ECCRN). (1996). Characteristics of infant child care: Factors contributing to positive caregiving. Early Childhood Research Quarterly, 11, 269–306. [Google Scholar]
  59. National Institute of Child Health and Human Development (NICHD) Early Child Care Research Network (ECCRN). (1998). Early child care and self-control, compliance, and problem behaviors at twenty-four and thirty-six months. Child Development, 69(4), 1145–1170. [PubMed] [Google Scholar]
  60. National Institute of Child Health and Human Development (NICHD) Early Child Care Research Network (ECCRN). (2000). The relation of child care to cognitive and language development. Child Development, 71, 960–980. [DOI] [PubMed] [Google Scholar]
  61. National Institute of Child Health and Human Development (NICHD) Early Child Care Research Network (ECCRN). (2006). The relations of classroom contexts in the early elementary years to children’s classroom and social behavior In Huston AC& Ripke MN(Eds.), Developmental contexts in middle childhood: Bridges to adolescence and adulthood (pp. 217–36). New York, NY: Cambridge University Press. [Google Scholar]
  62. National Institute of Child Health and Human Development (NICHD) Early Child Care Research Network (ECCRN), & Duncan GJ. (2003). Modeling the impacts of child care quality on children’s preschool cognitive development. Child Development, 74(5), 1454–1475. [DOI] [PubMed] [Google Scholar]
  63. Neuman SB, & Cunningham L (2009). The impact of professional development and coaching on early language and literacy instructional practices. American Educational Research Journal, 46, 532–566. [Google Scholar]
  64. OECD. (2013). Maternal employment rates: OECD Family Database, Social Policy Division, Directorate of Employment. Labour and Social Affairs, http://www.oecd.org/els/family/LMF1_2_Maternal_Employment_July2013.pdf [Google Scholar]
  65. Office of Child Care. (2014). Reauthorization of the Child Care and Development Fund (CCDF): An exciting new era for child care. Retrieved http://www.acf.hhs.gov/programs/occ/ccdf-reauthorization [Google Scholar]
  66. Office of Head Start. (2011). Designation renewal. Washington, DC: Office of Head Start Early Childhod Learning and Knowledge Center; Retrieved from eclkc.ohs.acf.hhs.gov/hslc/hs/grants/dr [Google Scholar]
  67. Peisner-Feinberg ES (2013). North Carolina pre-kindergarten program evaluation: Summary of research 2002–2013 Chapel Hill: University of North Carolina, FPG Child Development Institute; Retrieved October 15, 2014, from http://fpg.unc.edu/sites/fpg.unc.edu/files/resources/reports-and-policy-briefs/SummaryofNCPre-KEvaluationFindings2005–2014.pdf [Google Scholar]
  68. Peisner-Feinberg ES, & Schaaf JM (2007). Evaluation of the North Carolina More at Four prekindergarten program: Children’s outcomes and program quality in the fifth year. Chapel Hill, NC: FPG Child Development Institute. [Google Scholar]
  69. Peisner-Feinberg ES, & Schaaf JM (2008). Evaluation of the North Carolina More at Four prekindergarten program: Performance and progress in the seventh year (2007–2008). Chapel Hill, NC: FPG Child Development Institute. [Google Scholar]
  70. Pianta R, Howes C, Burchinal M, Bryant D, Clifford RM, Early DM, et al. (2005). Features of pre-kindergarten programs, classrooms, and teachers: Prediction of observed classroom quality and teacher-child interactions. Applied Developmental Science, 9(3), 144–159. [Google Scholar]
  71. Pianta RC, La Paro KM, & Hamre BK (2004). Classroom assessment scoring system [CLASS] manual: Pre-K. Baltimore, MD: Brookes Publishing. [Google Scholar]
  72. Pianta RC, Mashburn AJ, Downer JT, Hamre BK, & Justice L (2008). Effects of web-mediated professional development resources on teacher-child interactions in prekindergarten classrooms. Early Childhood Research Quarterly, 23, 431–451. [DOI] [PMC free article] [PubMed] [Google Scholar]
  73. Powell DR, Diamond KE, Burchinal MR, & Koehler MJ (2010). Effects of an early literacy professional development intervention on Head Start teachers and children. Journal of Educational Psychology, 102, 299–312. [Google Scholar]
  74. Preschool Curriculum Evaluation Research Consortium. (PCER). (2008). Effects of preschool curriculum programs on school readiness: Report from the preschool curriculum evaluation research initiative. Washington, DC: National Center for Education Research. [Google Scholar]
  75. Puma M, Bell S, Cook R, Heid C, & Lopez M (2005). Head Start Impact Study: First year findings. Administration for Children & Families. Retrieved October 15, 2014, from http://files.eric.ed.gov/fulltext/ED543009.pdf [Google Scholar]
  76. Raver CC, Jones SM, Li-Grining C, Zhai F, Bub K, & Pressler E (2011). CSRP’s impact on low-income preschoolers’ preacademic skills: Self-regulation as a mediating mechanism. Child Development, 82, 362–378. [DOI] [PMC free article] [PubMed] [Google Scholar]
  77. Ritchie S, Weiser B, Kraft-Sayre M, & Howes C (2001). Emergent academics snapshot scale (Unpublished instrument) Los Angeles: University of California. [Google Scholar]
  78. Rubin DB (1976). Inference and missing data. Biometrika, 63, 581–592. [Google Scholar]
  79. Rubin DB (1987). Multiple imputation for nonresponse in surveys. New York, NY: Wiley. [Google Scholar]
  80. Sabol TJ, Soliday Hong SL, Pianta RC, & Burchinal MR (2013). Can rating pre-k programs predict children’s learning? Science, 23(341), 845–846. [DOI] [PubMed] [Google Scholar]
  81. Schafer JL, & Graham JW (2002). Missing data: Our view of the state of the art. Psychological Methods, 7(2), 147–177. [PubMed] [Google Scholar]
  82. Schweinhart LJ, Montie J, Xiang Z, Barnett WS, Belfield CR, & Nores M (2005). Lifetime effects: The high/scope perry preschool study through age 40. Ypsilanti, MI: High/Scope Press, http://www.highscope.org/Research/PerryProject/perrymain.htm [Google Scholar]
  83. Shadish WR, Cook TD, & Campbell DT (2002). Experimental and quasi-experimental designs for generalized causal inference. Boston: Houghton Mifflin. [Google Scholar]
  84. Smith M, & Dickinson DK (2002). User’s guide for the early language and literacy classroom observation toolkit. Boston, MA: Brookes Publishing. [Google Scholar]
  85. Smith MW, Brady JP, & Anastasopoulos L (2008). Users guide to the early language and literacy classroom observation tool, pre-k. Baltimore, MD: Brookes Publishing. [Google Scholar]
  86. Sylva K, Stein A, Leach P, Barnes J, & Malmberg LE (2011). Effects of early child-care on cognition, language, and task-related behaviours at 18 months: An English study. British Journal of Developmental Psychology, 29(1), 18–45. [DOI] [PubMed] [Google Scholar]
  87. Tout K, Starr R, Isner T, Cleveland J, Albertson-Junkans L, Soli M, et al. (2011). Evaluation of parent aware: Minnesota’s quality rating and improvement system pilot. Minneapolis, MN: Child Trends. [Google Scholar]
  88. Tout K, Starr R, Soli M, Moodie S, Kirby G, & Boller K (2010). The Child Care Quality Rating System Assessment: Compendium of quality rating systems and evaluations. Office of Planning, Research and Evaluation, Administration for Children and Families, U.S. Department of Health and Human Services. [Google Scholar]
  89. Tout K, Zaslow M, Halle T, & Forry N (2009). Issues for the next decade of quality rating and improvement systems (OPRE Issue Brief #3). Washington, DC: U.S. Department of Health and Human Services, Administration for Children and Families, Office of Planning, Research and Evaluation. [Google Scholar]
  90. Tran H, & Weinraub M (2006). Child care effects in context: Stability and multiplicity in nonmaternal child care arrangements during the first 15 months of life. Developmental Psychology, 42, 566–582. [DOI] [PubMed] [Google Scholar]
  91. Departments of Education US and Health and Human Services. (2104). http://www.ed.gov/news/press-releases/six-states-awarded-race-top-early-learning-challenge-rtt-elc-grants-build-statew [Google Scholar]
  92. Vandell DL, Belsky J, Burchinal M, Steinberg L, Vandergrift N, & NICHD Early Child Care Research Network. (2010). Do effects of early child care extend to age 15 years? Child Development, 81, 737–756. [DOI] [PMC free article] [PubMed] [Google Scholar]
  93. Vermeer HJ, Groeneveld MG, Larrea I, van IJnendoorn MH, Barandiaran A, & Linting M (2010). Child care quality and children’s cortisol in Basque Country and the Netherlands. Journal of Applied Developmental Psychology, 31, 339–347. [Google Scholar]
  94. Vitiello VE, Moas O, Henderson HA, Greenfield DB, & Munis PM (2012). Goodness of fit between children and classrooms: Effects of child temperament and preschool classroom quality on achievement trajectories. Early Education & Development, 23, 302–322. [Google Scholar]
  95. Vogel C, Brooks-Gunn J, Martin A, & Klute MM (2013). Impacts of Early Head Start participation on child and parent outcomes at ages 2, 3, and 5 In Love JM, Chazan-Cohen R, Raikes H, & Brooks-Gunn J (Eds.), What makes a difference: Early Head Start evaluation findings in a developmental context. Boston, MA: Wiley. [DOI] [PubMed] [Google Scholar]
  96. Votruba-Drzal E, Coley RL, & Chase-Lansdale PL (2004). Child care and low-income children’s development: Direct and moderated effects. Child Development, 75, 296–312. [DOI] [PubMed] [Google Scholar]
  97. Wasik BA, & Hindman AH (2011). Improving vocabulary and pre-literacy skills of at-risk preschoolers through teacher professional development. Journal of Educational Psychology, 103, 455–469. [Google Scholar]
  98. Watamura SE, Phillips DA, Morrissey TW, McCartney K, & Bub K (2011). Double jeopardy: Poorer social-emotional outcomes for children in the NICHD SECCYD experiencing home and child-care environments that confer risk. Child Development, 82 (1), 48–65. [DOI] [PubMed] [Google Scholar]
  99. Weber R (2005). Measurement of child care arrangement stability: A review and case study using Oregon Child Care Subsidy Data. (Doctoral dissertation). Spokane: Oregon State University. [Google Scholar]
  100. Weiland C, Ulvestad K, Sachs J, & Yoshikawa H (2013). Associations between classroom quality and children’s vocabulary and executive function skills in an urban public kindergarten program. Early Childhood Research Quarterly, 28, 199–209. [Google Scholar]
  101. West J, Tarullo L, Aikens N, & Hulsey L (2008). Study design and data tables for FACES 2006 Baseline Report. Washington, DC: U.S. Department of Health and Human Services. [Google Scholar]
  102. Woodcock RW, & Johnson MB (1989). WJ-R tests of cognitive ability. Itasca, IL: Riverside. [Google Scholar]
  103. Woodcock RW, McGrew KS, & Mather N (2001). Woodcock–Johnson III: Tests of achievement. Itasca, IL: Riverside. [Google Scholar]
  104. Yamauchi C, & Leigh A (2011). Which children benefit from non-parental care? Economics of Education Review, 30, 1468–1490. [Google Scholar]
  105. Zaslow M, Crosby D, & Smith N (2014). Research issues related to early care and education emerging from the changing policy landscape In Gershoff E, Mistry R, & Crosby D (Eds.), Societal contexts of child development: Pathways of influence and implications for practice and policy (pp. 54–74). New York, NY: Oxford University Press. [Google Scholar]
  106. Zaslow M, Martinez-Beck I, Tout K, & Halle T (Eds.). (2011). Quality measurement in early childhood settings. Baltimore, MD: Brookes Publishing. [Google Scholar]
  107. Zaslow M, Tout K, Halle T, Whittaker JV, & Lavelle B (2010). Toward the identification of features of effective professional development for early childhood educators. Washington, DC: U.S. Department of Education, Office of Planning, Evaluaiton and Policy Development, Policy and Program Studies Service; http://www.ed.gov/about/offices/list/opepd/ppss/reports.html [Google Scholar]
  108. Zellman GL, Perlman M, Le V, & Setodju CM (2008). Assessing the validity of the qualistar early learning quality rating and improvement system as a tool for improving child care quality (MG-650-QEL) Santa Monica, CA: RAND Corporation. [Google Scholar]
  109. Zill N (1990). Behavior problems index based on parent report [memorandum] Washington, DC: Child Trends. [Google Scholar]

RESOURCES