ACF OPRE: Head Start Impact Study and Follow-up

Chapter 6: Impact of Head Start on Children’s Social-Emotional Development

Highlights

Among children in the 3-year-old group, those in the Head Start group were reported by their parents to show less frequent and severe problem behavior than non-Head Start children in the same age group. However, the magnitude of the positive impact of Head Start on children’s social and emotional development, while statistically significant, was relatively small. No overall statistically significant impacts for social-emotional outcomes were found for children in the 4-year-old group overall, but impacts were found for the subgroup of children from English-speaking families. Specifically:

Positive impacts of Head Start were found on the Total Problem Behavior measure for children in the 3-year-old group (effect size of 13 percent of a standard deviation) and the Hyperactive Behavior measure for the same age group (effect size of 18 percent of a standard deviation).
A positive impact was also found on the Aggressive Behavior measure for children in the 4-year-old group, but only for those children from English-speaking families (effect size of 15 percent of a standard deviation).
No overall impact of Head Start was found on parent-reported Social Skills and Positive Approaches to Learning of children, nor on parent-reported Social Competencies, for children in both age groups.

It is important to note that the analysis of Head Start’s impact on children’s social and emotional development has thus far relied solely on behavior reports from parents. An important additional source of information on children’s social development and positive and negative behaviors, reports from children’s teachers and caregivers, could not be used at this stage of the impact analysis. This is because such reports on non-Head Start children who only had parent care are not available. Teacher reports on the social behavior of all study children will be available in future years of the study, when the children are in elementary school.

Organization and Presentation of Findings

This chapter focuses on the impact of Head Start on the following three constructs of children’s social-emotional development:

Social skills and approaches to learning;
Social competencies; and
Problem behavior (total problem behavior, aggressive, hyperactive, and withdrawn).

As in Chapter 5, the discussion of estimated impacts focuses on an examination of statistically significant “intent-to-treat” impact estimates using the complete sample of children who were randomly assigned in 2002. The discussion begins with the overall average impacts for all newly entering children in both the 3- and 4-year-old groups, and then examines any notable differences in average impacts by the language used for child assessment. Finally, the discussion moves to an examination of the extent to which impacts occurred for key subgroups of Head Start children, and how different in size impacts may be for various subgroups. As in Chapter 5, estimated impact of Head Start on program participants is provided in Appendix 6.1, examining the combined group of all children (i.e., separate breakdowns by the language of assessment are not provided).

The statistical results for the discussion in this chapter are presented in a series of tables appearing at the end of this chapter and in Appendix 6.2. Exhibits 6.1-A through 6.1-C (for children in the 3-year-old group) and 6.2-A through 6.2-C (for the 4-year-old group), present the overall average impact estimates for the combined sample and for the two separate language groups. As in Chapter 5, the data are presented in three ways: (1) as simple mean differences, (2) using regression analyses that include only demographic covariates measured in fall 2002, (3) and using regression analyses that added a measure of the outcome variable assessed in fall 2002. Shading is used to indicate which estimate is discussed in the text. Also similar to Chapter 5, Exhibits 6.3-A and 6.3-B (for the 3- and 4-year-old groups, respectively) summarize all of the statistically significant impacts (both for the overall group and for a set of 15 subgroups discussed in Chapter 4) and provide the average impact of access to Head Start along with associated effect sizes. Finally, Exhibits 6.4 through 6.15, provided in Appendix 6.2, show the results of the moderator/subgroup analyses, with a separate table for each individual measure of social-emotional outcomes (again, only for the full combined sample).

Estimated Impact of Access to Head Start

This first section discusses the estimated impact of Head Start on social-emotional outcomes using the sample of children randomly assigned to either Head Start or to the non-Head Start group, referred to as “intent-to-treat” impact estimates. These estimates show the effect of Head Start on the average child given access to the program.

Impact on Social Skills and Approaches to Learning

As reported by parents, the Social Skills and Positive Approaches to Learning (SSPAL) of children and the Social Competencies Check List (SCCL) in both the 3- and 4-year-old groups did not show an impact of Head Start, i.e., the scores on the SSPAL and the SCCL scales did not statistically differ between children in the Head Start group compared to children in the non-Head Start group.

Impact on Problem Behavior

As shown in Exhibit 6.1-A for children from all language groups in the 3-year-old group, the score on the Total Behavior Problems scale at the end of the program year was 0.5 points lower for children in the Head Start group compared to children in the non-Head Start group. The score on the Hyperactive Behavior subscale of the Total Behavior Problems scale was also significantly lower by 0.3 points for Head Start children compared to the non-Head Start children.

No overall statistically significant impacts were found for these outcomes for children in the 4-year-old group (Exhibit 6.2-A). However, among children in this age group from English-language family backgrounds (Exhibit 6.2-B), children in the Head Start group scored significantly lower on the Aggressive Behavior subscale at the end of the program year than did similar children in the non-Head Start group (a difference of 0.22 points).

Although statistically significant, the magnitude of the Head Start impact on children’s problem behaviors was small (see effect sizes in Exhibits 6.3-A and 6.3-B). The effect sizes of the impact on Total Problem Behavior of children in the 3-year-old group was 13 percent of a standard deviation, and for the impact on Hyperactive Behavior, the effect size was 18 percent of a standard deviation. The effect size of the impact on the Aggressive Behavior of 4-year-olds from English-speaking families amounted to 15 percent of a standard deviation.

Moderator/Subgroup Differences

The analysis of impacts by subgroups of children and families (detailed in Appendix 6.2 and summarized in Exhibits 6.3-A and 6.3-B for those found to be statistically significant) show some variations in impact for particular types of Head Start participants. The most important, and consistent, of these findings are discussed below, related to impacts on particular subgroups. The subgroup-specific impact findings for social-emotional outcomes, although less widespread than for the cognitive outcomes, show that children from different language and racial/ethnic groups are benefiting from Head Start.

Other statistically significant findings in Exhibits 6.3-A and 6.3-B are not discussed because it is possible they are due to chance alone and do not represent true impacts of the intervention (see discussion of subgroup impact analysis in Chapter 4).1

Impacts on Particular Subgroups

Child language. Among children in the 3-year-old group, significant negative impacts of Head Start on Total Problem Behavior were found for the subgroup of children from English-speaking families (i.e., Head Start reduced the incidence of reported problem behaviors), while significant negative impacts on Hyperactive Behavior were found separately for children from English-speaking families and for children from Spanish-speaking families (i.e., reductions in the reported incidence of aggressive behavior). Among children in the 4-year-old group, the significant negative impact of Head Start on Aggressive Behavior was found only in the subgroup of children from English-speaking families. No other differences were found to be statistically significant for either language group.

Child race/ethnicity. Among children in the 3-year-old group, there is a negative impact of Head Start on the reported Social Competencies Checklist, i.e., the parents of African American children in the Head Start group reported less social development for their children than did parents in the non-Head Start group. For Hispanic and White children in this same age group, positive impacts were found in the area of Hyperactive Behavior (i.e., a reported lower incidence of such behavior). Among African American children in the 4-year-old group, significant impacts of Head Start were found on reported Total Problem Behavior and on the reported Aggressive Behavior subscale (i.e., indicating reductions in reported negative behaviors).2

Exhibit 6.1-A: Initial One-Year Estimates of the Impact of Access to Head Start on Social-Emotional Outcomes: 3-Year-Old Group, Combined Fall English-Spring English and Fall Spanish-Spring English Group (Weighted Data)
Outcome Measure	Intent-To-Treat Impact Estimates
Outcome Measure	Head Start Mean	Non-Head Start Mean	Mean Difference 2	Regression-Adjusted Impact Adjusted Impact Estimates (Demographic Covariates Only)	Regression-Adjusted Impact Estimates (With Fall Measure)
(Sample N=2,071):
Social Skills Scale	12.4	12.4	0.1	0.06	-0.0
Total Problem Behavior Scale	5.8	6.3	-0.5*	-0.49*	-0.48** (-0.13)
Aggressive Behavior Scale	3.0	3.0	-0.1	-0.07	-0.12
Hyperactive Behavior Scale 1	1.7	2.0	-0.3**	-0.34***	-0.29** (-0.18)
Withdrawn Behavior Scale	0.6	0.6	-0.0	-0.03	-0.03
Social Competencies Checklist	11.0	11.0	-0.0	-0.04	-0.06