ECONOMIC OPPORTUNITY STUDIES

400 NORTH CAPITOL STREET, SUITE G-80, WASHINGTON, D.C. 20001

Tel. (202) 628 4900 Fax (202) 393 1831 E-mail info@opportunitystudies.org

_________________________________________________

Head Start Programs and Services:

Does H.S. Local Agency Type Make a Difference to Children?

December 15, 2001

Summary

Study of Head Start (H.S.) program elements and National Head Start Bureau performance indicators in each of the five types of local agencies administering FY 2000 Head Start programs for three to five year olds showed few differences among agency types. This analysis examined the indicators of services provided by Community Action Agencies (CAAs) compared to other types of Head Start delivery agencies and found their performance to be close to, or higher than, the national average.

Study of the family services and other non-educational services provided by CAAs showed them to be similar in scope and staffing to those in the private non-profit agencies, but better staffed and more comprehensive than those in the school-based or government agency Head Start agencies.

• CAAs had staffing ratios, including family caseworker-to-family ratios at, or better than, the national average;

• CAAs were the type of agency most likely to find and serve homeless children;

• CAAs had a high proportion of credentialed classroom staff, comparable only to that in school system programs;

• All types of agencies performed well on the Bureaus national measures according to their own FY 2000 reports.

Introduction - Economic Opportunity Studies, Inc. (EOS) of Washington, DC examined national Head Start (H.S.) program data from FY 2000 to identify whether Head Start services delivered by Community Action Agencies were unique compared to other types of local H.S. grantees. The service was measured by program and family indicators reported to the H.S. Bureau in the annual local Program Information Reports¹. This review was commissioned by the National Association for State Community Services Programs (NASCSP), with funding from the United States Department of Health and Human Services (USDHHS) Office of Community Services (OCS).

The resulting comparisons were not intended to substitute for performance measurement nor for detailed evaluation of outcomes for the five categories of local H.S. agencies. At best, it could suggest a relationship between the nature of the sponsoring institution and some characteristics of programs, an association that would require detailed analysis to find five predictors of different outcomes. The report is intended for use by at least three partners in contemporary initiatives to strengthen H.S. programs:

- The Office of Community Services, which is now supporting extensive technical assistance activities for local agencies Head Start programs and is also engaged in the CSBG network implementation of a Results-Oriented Management Assessment system, and by

- The joint national and state working groups established by the OCS and the Head Start Bureau (HSB) to improve CSBG/HS collaborative strategies (as per M.O.U. 8/26/98); and

- The H.S. national network of local agencies

Agency Type: Potential Policy Issues

The question of whether different types of Head Start (H.S.) agencies perform at all differently with respect to the services provided to H.S. children or their families is the central question explored here. The Community Services Block Grant network agencies, or Community Action Agencies (CAAs), are nearly all multi-service, community-based organizations and have as their core purpose the elimination of the causes of poverty by means of integrating, or at least identifying, multiple resources for families and communities. Many Indian tribal programs are also multi-purpose. There is extensive data on the activities and resources of the 1,120 CAAs in the CSBG Information System (CSBG/IS), based on annual reports collected and maintained by the National Association for State Community Services Programs (NASCSP) with support from the U.S. Department of Health and Human Services, Office of Community Services². Over half of all CAAs are included here. When combined with reports from the CAAs that are contained in the H.S. Program Information Reports (PIR), the nature and extent of services the CAA network made available to FY 2000 H.S. participants can be described generally. Some of the private non-profit H.S. agencies are similar to CAAs in purpose if not in size while others are predominantly early childhood program centers. School systems are, obviously, education specialists. Agencies self-reported types were verified against CSBG subgrantee lists and modified where necessary. All changes made are shown in Appendix 1.

Scope of the Study

This analysis used only Head Start PIR data from the 50 states and the District of Columbia (DC), and from Puerto Rico, the jurisdictions that participate in the CSBG/IS. Early Head Start program information was not part of the analysis. As a result, the national totals, although taken from the FY 2000 PIR, are smaller than those in the Head Start Bureaus national reports. The FY 2000 PIR covered 2,431 delegate agencies, including Early Head Start providers with a funded enrollment total of 843,982. This studys population includes 1,911 delegate agencies with 797,119 total funded enrollments in classes, centers, or home-based programs. As compared to combined totals reported for all H.S. and Early H.S., this group represents 79 percent of the agencies, and 91 percent of the enrollment in H.S. and Early H.S. combined.

Table 1

The Scope of the Program by Type of Delegate Agency:

*Percentages do not add due to rounding

**This is the average daily enrollment for Nov. 1999, February 2000, and March 2000.

Table 1 above shows that more than two-thirds (69 percent) of the H.S. agencies were either CAAs or private non-profit agencies. CAAs constituted about a third. Together, these two types of providers managed over three-fourths of all the H.S. enrollments funded from any source.

Table 2 illustrates the localized interconnection among different kinds of H.S. agencies. CAAs are the one type of agency most likely to pass through, or delegate, H.S. grants to other local agencies. More than 10 percent of all local agencies were not CAAs, but received sub-contracts from CAAs that were H.S. grantees. Of those 196 sub-delegates, 103 were private non-profit agencies.

Table 2

Number of Sub-Delegate Agencies who contracted

with CAA delegate agencies, by type:

With enrollment in the sub-delegates is considered, nearly 312,000 H.S. participants, over 39 percent of the FY 2000 total, were served by CAAs or by agencies to which CAAs had delegated the FY 2000 program. The remainder of this analysis examines only the programs and services delivered by each type of delegate agency regardless of the source of grants or sub grants.

Does agency type make a difference?

First, we examined three sets of variables related to the nature and quality of H.S. programs. Class size, staff resources, and services available to participants and their families, as measured annually by the HSB Bureau.

The results of all these comparisons are shown as an appendix to the report. The analysis covers only the factors on which there is variance of 10 percent or more between the CAAs and/or other providers by type.

H.S. Classes and Staffing

On several dimensions of program indicators CAAs are essentially at the H.S. national average. The size of their classes and the average agency ratio of classroom; center, and agency staff per child, are indistinguishable from other kinds of agencies. Only Indian tribal programs have notably lower adult-to-child ratios and smaller classes. Table 3 shows these data in the three left-hand data columns.

Table 3

Staffing Ratios by Agency Type:

Type of Agency	Kids Per Class (i.e. Average Class Size)	Kids Per Agency Staff	Kids per Classroom / Center Staff	No of Families (i.e. Caseload) per Caseworker
CAAs	18	4.8	9.4	40
Government	19	4.5	9.4	51
Indian Tribe	17	4.1	7.4	20
Non-Profit	19	4.8	9.4	40
School System	18	5.7	9.6	46
All Agency Types	19	4.9	9.4	40

On one dimension of staffing resources, CAAs and non-profits do differ significantly from schools and other agencies. Only Indian Tribes have more family casework staff. Measured by the total families served at any time in the year divided by the number of agency staff whose primary responsibility is working with families, CAAs and non-profits have enough such staff to keep the caseload to an average of 40 H.S. families a year. The School Systems average is 15 percent higher and government agency family workers have 20 percent higher loads. This appears to be a clear reflection of the different roles the five kinds of agencies play with regard to their overall mission; the networks consisting predominantly of multi-service organizations that serve families and communities have more trained staff for that purpose.

Table 4 Percentage of Agencies that Serve Homeless Children and Their Families

Average by Agency Type

Type of Agency	Percentage of Agencies that Undertake Special Initiatives to Serve Homeless Children and Their Families
CAAs	67%
Government	46%
Indian Tribe	36%
Non-Profit	56%
School System	55%
All Agency Types	47%

In another, but related, dimension of family services, CAAs appear to be as disproportionately more likely to serve homeless children and families, as shown in Table 4 below. The FY 1999 CSBG/IS suggests the reasons. CAAs collectively expended at least $100 million of state and federal programs for the homeless in FY 1999; further, a major element of their services involved responding to low-income families with emergency needs. About 18 percent of CAAs core CSBG funding was devoted to Emergency Services that were coordinated with other specific programs³.

One other factor on which there is significant variation among agency types is in the ethnic and racial mix of their clients. Table 5 below shows the details. CAAs enroll 37 percent of H.S. participants and serve a majority of all white children in the H.S. program. Because of their role in Hawaii, where most Pacific-Islander children live, CAAs also serve a majority of that population. The CAA share of African-American enrollees is just slightly higher than their share of all U.S. participants; they enrolled 42 percent of that group, or a little more than the number served by non-profit agencies. Hispanic children, 29 percent of the total, are most likely to be at private non-profit H.S. agencies, while CAAs enroll only 19 percent of them.

Table 5

Percent of H.S. Participants of Each Ethnic / Race group served by Agency Type.

Close examination shows that these demographics are closely related to the geographic distribution of CAAs. More than half of all Hispanic H.S. participants, 56 percent, lived in either California, Texas, or Puerto Rico, yet CAAs represented a small fraction of the H.S. agency network in California and Texas, with 8 percent and 9 percent of enrollment respectively. They played no role in the Puerto Rico H.S. program. Although Hispanic participants did make up two-thirds of all the children the CAAs served in their California and Texas H.S. programs, the impact on the significant population of eligible Hispanic children is not large. In fact, the ethnic variation among the agency types is very likely to be related to the local or state history that made one or another kind of community organization the H.S. grantee. Nevertheless, when taken together with CSBG/IS client data for the FY 1999, which showed that Hispanic clients made up just 14 percent of those served by CAAs, these data support future use of CAA resources to broaden the reach of services in Hispanic low-income communities.

In a previous report, NASCSP noted the regional variation in the scope of CAA H.S. services⁴. In regions 8 and 9, for example, CAAs serve a smaller share of participants than in regions 1 and 5.

The H.S. Performance Indicators

The H.S. Bureau reports annually on a selected set of performance indicators by program state and region. We have adapted that report to compare the five types of agency in the H.S. program including Indian, and Migrant, but not Early, Head Start. These are attached as Appendix 3, which shows CAAs are close to or above national norms on all these key standards.

Other Family and Program Characteristics

Appendix 2 shows the other program and family variables that did not differ significantly by agency type. Its tables 1-4 show variables related to the scope and nature of the Head Start program. While CAAs had a proportionate share of the centers, all kinds of agencies ran the program predominantly through 5-day center-based classrooms. No single agency type had a significantly larger share of Locally Designed or Home-Based programs than another did.

The staff qualifications were similar by category for all kinds of agencies. CAAs and school systems had more classroom staff with some level of child development or teaching credential than did other agency types. In fact, CAAs were indistinguishable from school systems with respect to the credentialed proportion of their classroom personnel when all levels of credentials were considered. All types of agencies had volunteers roughly in proportion to their share of all enrollees, and two-thirds of the volunteers were H.S. parents.

Family characteristics were generally similar by agency type; CAAs H.S. clients were as likely to be from working poor or unemployed families as those served by other agencies. TANF recipients children were found in all kinds of agencies. All kinds of agencies, except schools, found about 27 percent of families had emergency needs, and all five types of agencies believed they had helped resolve the vast majority of such problems. Educational needs were identified in 22 percent of all families and most received services regardless of their H.S. agency type.

CAAs, non-profit agencies, and schools were more successful than government agencies and tribes at securing subsidized daycare for the children who needed it and about half was paid for with Head Start resources. Nevertheless, a majority of children in every kind of agency relied on care that was given to them for free or paid for by their family.

Conclusion

CAAs and their sub-delegates are responsible for services to nearly 40 percent of all children in the core Head Start program for 3-to-5 year olds. As a national network, as well as individually, they meet the HHS performance indicators as well or better than the other types of local agencies. As measured by services provided to the families participating and the quality of their staff. CAAs at least equal, and in several dimensions, exceed the work of other types of agencies.

FastCounter by bCentral