ACF OPRE CORE - Publications and Reports - Reaching Out to Families: Head Start Recruitment and Enrollment Practices

4.0 Conclusions and Lessons Learned

4.1 Overview

From casual observers of Head Start to Federal program staff to the Advisory Committee on Head Start Research and Evaluation, there has been an ongoing interest in learning more about eligible families who remain unserved by Head Start. This feasibility study was conceived with a similar intent, and specifically addressed two general goals. First was an effort to understand the current perspectives of Head Start staff regarding recruitment and enrollment activities, while the second goal was to assess the state of knowledge regarding enrolled and non-enrolled Head Start-eligible families. As part of this second goal, the study team sought to determine the feasibility of identifying and engaging the parents of non-enrolled, Head Start-eligible preschool-age children in individual interviews. A discussion of the key findings of this effort is presented below, followed by a listing of the major lessons learned from this project.

4.2 Discussion of Findings

Findings from the multiple data sources used in this study are discussed below to build a better understanding of how the Head Start program staff worked to bring families into the program and to determine what information is available about the eligible, unserved families.

Head Start Recruitment Activities. The recruitment strategies identified by programs were generally uniform across all the focus group sites. Head Start staff reported that recruitment was viewed as an on-going, year-round process, with the most intense recruitment periods being spring and summer, a finding supported by the listed enrollment dates noted in the recruitment record reviews. These records indicated that families were applying throughout the calendar year, with most applying during the few months prior to the beginning of the Head Start year. While many staff noted the use of formal recruitment activities (taking referrals from community agencies, setting up booths at community events), it is important to note that they felt that recruitment really occurred anytime the opportunity arose, even through activities as informal as observing and meeting a family with a preschool-age child in a grocery store in an area populated by eligible families. While this philosophy suggested that all staff persons, from the program administrators to teachers’ aides, had responsibilities in this area, the majority of the recruitment efforts were actually handled by the field staff. Staff from rural programs noted that their recruitment activities were managed at the center level, while the urban staff reported that recruitment was more of a program-wide effort. The reports from the staff focus groups on recruitment activities were similar to those of Head Start staff in previous work on this topic (Love & Grover, 1987).

Head Start staff, particularly the field staff, reported that while they tried to target the “neediest of the needy” in their recruiting, their recruitment efforts with these particular families were not always successful. They suggested that breakdowns in recruiting these families have, in turn, sometimes led them to recruit eligible, but less needy families. They indicated that this was done to ensure that the program met its targeted enrollment by specified dates. Field staff expressed concern about the potential loss of Federal funds and, therefore, the potential loss of their jobs, if these enrollment target dates were not met.

During recruitment, the field staff in many programs met with families to assist in the completion of applications, either in the homes of the families, at designated program sites, or during community outreach activities. The Head Start staff person used these opportunities to observe the areas of family need and review appropriate documentation for the verification of income to determine income eligibility for enrollment. Staff reported that some of the parents they recruited felt the level of required documentation for enrollment was too invasive.

Field staff noted that particular family situations may have influenced how far they actually would go to verify the income information provided by the families. For example, staff reported that they have sometimes encountered families with social service needs who did not qualify under the program’s income eligibility guidelines. In some of these cases, staff have accepted income information from families without full verification or took an older (and acceptable) piece of documentation that suggested the family was income-eligible, rather than more recent information that would have left the family over the income eligibility threshold and out of Head Start. Staff admitted that they occasionally “bent the rules,” reflecting their need to meet enrollment targets or, in the more likely case, highlighting their desire to help families who they believed were in particular need of Head Start services, but would not have qualified for Head Start under a strict application of the program’s income guidelines. In many cases, even a few over-income families would not have been a problem, given the leeway programs have to accept up to 10% of their enrollment with incomes over the FPL. However, the tone of this particular discussion by the staff clearly implied that they felt they were not strictly adhering to the guidelines they were given for recruitment.

A number of factors were reported that impacted local recruitment efforts. The staff focus groups noted that their recruitment activities have been influenced by the mobility of families. This was consistent with Love and Grover’s report in 1987 about pockets of eligible families that moved within a program’s service area while seeking work and affordable housing. Current staff also noted that family mobility, for the same reasons, impacted their recruitment activities and their ability to offer Head Start services to families in parts of their service areas where Head Start was not prepared to serve (e.g. based on the proximity of centers and the availability of transportation). The general availability of transportation in an area sometimes factored into decisions to recruit particular families in particular areas, because these children were not able to be transported to class each day by the program.

Administrative staff acknowledged the importance of matching the cultural or ethnic background of their staff with the families Head Start was seeking to serve, but they admitted that this was not often the practice. Field staff concurred, reporting that the task of recruiting new families from different ethnic groups was a problem when the staff sent to those families was not representative of that ethnic group. This was particularly true in areas where communities of immigrant families have grown quickly, often in response to new work opportunities. Unfortunately, while some recruitment staff may have been able to speak the native language of these families, they were less likely to understand the culture. Many staff recognized the importance of learning about the culture of families in their efforts to build links to these new communities, and felt unprepared to recruit and serve these potentially needy families.

During recruitment activities, staff noted that Head Start was coming face-to-face with new competition for 4-year-old children from both public preschool programs and state-subsidized child care. Staff felt that some parents found these alternatives superior to Head Start, for reasons that ranged from the increased educational focus of these options to the increased convenience families found in having all their children at one school during the day. These reports were supported by the parent focus groups, where educational issues were a primary focus of many non-Head Start parents. One point that did become clear was that despite the best intentions of the local Head Start staff, there remained a mismatch between local program models (full- or part-day, home- or center-based) and the needs of the families in some locations.

In focus group discussions, staff, particularly the administrative staff, acknowledged the formality of the established boundaries between the service areas of neighboring Head Start programs, yet reported that these boundaries were often “blurry.” When service boundary lines were in conflict, staff reported they were more than willing to work with another Head Start program to problem solve for solutions to help specific families that may have been unserved. For example, in some rural areas, centers in one program were used to serving children from an adjacent service area that had no centers near that location. In another example, one of the urban programs jointly sponsored a center with another local Head Start program near the shared border of their service areas.

Just as researchers have often focused on barriers to low-income families’ receipt of services, Head Start staff also encountered barriers in their recruitment activities. These barriers included 1) addressing families’ misconceptions about Head Start (who is eligible, what the program offers), 2) the presence in the communities of alternative services (public preschool, subsidized child care) that may better serve families’ needs, and 3) an inability on the part of the program itself (lack of physical space, transportation, inadequate hours, or specialized staff) to reach or serve all the families that could benefit from its services. In general, the recruitment activities across the programs showed that Head Start staff were trying to make inroads into the communities of families that needed them most, but they were not always able to serve these families. What these staff, particularly the field staff, brought to the effort to combat these barriers was the passionate belief that Head Start provides a superior service for children and families, and the willingness to go to considerable lengths to bring needy families into the program.

Head Start Enrollment Activities. There was little consistency across programs in the actual process of selecting families, but all program staff suggested they had the opportunity to adapt or override the formal system in order to serve particular families, when needed. The final decision for selecting families for enrollment was left to an individual or to some form of committee, and the make-up of these committees varied from program to program. All the committees included staff, and some also included parents from the Parent Advisory Committee.

Common to the enrollment decision was the use of predetermined eligibility risk factors to assess family need. Because these risk factors were selected to reflect the needs or risks of the communities where these families lived, they also varied from site to site. In most, but not all cases, these risk factors carried assigned values that were summed to generate a priority score for each family. The higher the score, the greater the risk for that family and the more likely they were to be enrolled in the program. The use of local community assessments to establish enrollment objectives was consistent with the findings of Love and Grover (1987). It was interesting, however, to learn that even after objective priority scores were determined for families, in most programs the recruitment staff – the individuals who assisted families with the application process – usually had opportunities to subjectively advocate for particular families they thought would benefit from the program. Advocacy was based on direct observations of family needs that staff felt were not reflected in their final priority score based on the compilation of eligibility risk factors.

Once all available classroom slots were filled, the remaining families were placed on a waiting list. More often than not, the waiting lists were kept at the individual center-level, but staff did report that occasionally, when spaces opened up, they would offer the spot to children from other centers within their program if these families were more needy (had higher priority scores) and if transportation was available. Unfortunately, the issue of waiting lists was somewhat frustrating for staff. The PIR findings suggested that there is a need for waiting lists based on the number of families that typically dropped out of programs during the course of a year. The FACES staff interviews offered strong evidence of great variation in the size of these waiting lists, both across and within programs. Head Start staff suggested in the focus groups that many families who were put on the waiting list would never actually enter the program. These families usually sought alternate sources of care for their children, citing that they could not wait for an opening in Head Start that might never come their way.

Head Start Retention Activities. During recruitment and enrollment activities, staff encountered families who quickly chose not to come to Head Start. Other families enrolled and started the program, but then chose to withdraw. It is important to understand why families left Head Start and how program procedures have addressed this issue. A review of the focus groups found similar follow-up procedures were used across all programs when children stopped coming to Head Start. These procedures included formal (letters) and informal (home visits and telephone calls) contacts with families after a child had a series of consecutive, unexplained absences (typically three consecutive days). What staff reported was that families left Head Start for a variety of reasons. In order of frequency, these reasons included families moving (and often not notifying Head Start), problematic family situations that precluded the family from having the child at the program (domestic problems, substance abuse, mental illness, or lack of motivation, organizational skills, or coping skills), the failure of Head Start to offer certain needed services (full-day care), separation issues for parents with young children, and transportation difficulties.

These findings were confirmed by the FACES exit interviews and the parent focus groups, which also provided some evidence of dissatisfaction with the local Head Start programs that was not typically evident in the FACES data. In fact, for the eight program satisfaction items used in both the FACES parent interviews and the exit interviews, the frequency of dissatisfaction ranged from 2.5 to 7 times greater among parents who had left the program (see Exhibit 2.8). In general, dissatisfaction was still reported by a relatively low percentage of parents, with less than 20% of parents reporting dissatisfaction on any of the eight items, and on a majority of the items, less than 10% of parents noted dissatisfaction. Similar to the reported findings in Love and Grover (1987), it also appeared that the reasons families left Head Start were very similar to the reasons other families chose not to enroll in Head Start in the first place. Staff indicated that they worked hard to retain families wherever possible, and that by working with these families on problem solving and creating solutions, Head Start often became more attractive and viable.

Perceptions of Head Start. One area noted as having significant impact on recruitment, enrollment, and retention was the perception of Head Start that was held by families and by the community agencies serving low-income families in each particular location. The FACES interviews conducted with community agencies found that while most agency representatives reported positive relationships with Head Start, some reported that program staff were unwilling to collaborate with them on activities that would serve their shared target population. Less than one half of the agencies contacted indicated that they regularly made referrals to Head Start. Some agency staff reported that there was an “elitist attitude” on the part of the local Head Start program that made it difficult for Head Start to actively develop the links it needed with other community agencies. Interestingly, this view was echoed by one of the administrative staff focus groups. This group reported that Head Start was perceived as arrogant by the local child care community, that Head Start looked on them with disdain, and would make unfair generalizations about the quality of care in non-Head Start settings. These administrators suggested that a challenge for Head Start would be to build better links with the informal child care network in their communities. These represent the type of links that Head Start needs to reach eligible, unserved families as well as to provide families with access to needed services.

Some parents reported concerns about the perceived quality of a Head Start education. In at least one location, parents and staff reported that children were labeled as “Head Start” children in the local school system, reflecting the negative perception of the school staff towards these children. Parents also felt that Head Start sacrificed education for socialization, and some Head Start staff felt they were viewed as unqualified educators or glorified babysitters. While the reports came primarily from the parents and staff at a few programs, the notion that Head Start was a play program without an educational component or plan was certainly not unique to those sites.

Misconceptions about the program were noted by staff and actually demonstrated by parents during the individual interviews. Eligible parents who had no experience with the program and did not know other families who were enrolled likely based their enrollment decisions on what they understood about the program, including reported misinformation about Head Start eligibility and what the program actually provided in terms of services for low–income children and families. Misconceptions were as simple as thinking that Head Start was a program that served only working families, children with behavioral problems, minorities, or disabled children. Successful recruitment efforts require staff to engage these families and communicate with them so accurate pictures of how Head Start serves low-income families can be fostered and families do not overlook opportunities that may provide them with critical benefits. Community consciousness-raising about Head Start was discussed by some staff in their focus groups as a method for improving the image of Head Start in the community and helping bring families from the target population into the program.

Identifying Characteristics of Eligible, Unserved Families. The exercise of reviewing national datasets for information on enrolled and non-enrolled Head Start-eligible families did not yield many conclusive findings. While the depth of information on eligible families was slim, one clear conclusion was reached. All the relevant data sources confirmed that nationally there have been large numbers of Head Start-eligible families who were not enrolled in the program. This conclusion is similar to one proposed by Love and Grover (1987) and is supported more recently by the findings of Nord (1999), as well as by the perceptions of the Head Start administrative and field staff during the focus group discussions. Even in the program sites considered to be “fully served,” staff acknowledged that pockets of eligible, unserved families existed.

The large national datasets (e.g. SIPP, NLSY79, PSID) seemed to suggest that Head Start served a number of families who were identified as over-income. This may be due to a number of factors, such as the natural maturation of some families, the potential impact of welfare reform on increasing family income, as well as the potential positive impact received from Head Start participation. It also may be that Head Start staff consciously or unconsciously make exceptions to the income guidelines for some of these families. One consistent finding across all the administrative and field staff focus groups was that staff regularly encountered families who needed Head Start services but were barely over the income-eligibility threshold. Just as Love and Grover (1987) found in the pre-welfare reform era, staff strongly believed that Head Start was missing an important opportunity to assist families in need. They emphasized that the need they saw was not satisfied by simple economic improvements. In fact, some needs grew as families became ineligible for certain services with the increase in their family incomes. The focus groups with administrative and field staff invariably spurred passionate discussions about the need to adjust income guidelines upwards, arguing that the needs of many families living on incomes above the traditional eligibility cutoff (Federal Poverty Level) were as great as those for families considered to be living in poverty.

Staff also suggested that unserved families in their communities may be those who lacked the necessary knowledge or means to access the local child and social service networks, as well as families who simply chose not to use services of any sort, Head Start or otherwise, preferring to manage on their own. This latter group included families who were just more comfortable having their children stay at home or with a family friend until they started kindergarten. Some families liked the comfort and informality of home and family day care settings and chose to forego the opportunities that Head Start might bring them.

Much of what was reported here on unserved families came directly from staff reports and the pilot parent interviews, and not from the more quantitative national data of the SIPP, NLSY79, and PSID. As noted earlier, the differences in the construction across the national datasets were serious enough to preclude their use in generating a consistent picture of families who were not in Head Start. These datasets also lacked the necessary information to offer insight into what caused families to not enroll in Head Start when they were eligible. For some families, the reason was simply the fact that they were unaware of the program, but for others there was a clear choice not to participate. Given that the national datasets did not address these issues and did not provide a consistent picture of the risks faced by these families, they were not sufficient to provide the necessary information noted at the start of this section. Further investigation is warranted, as long as there is a need for this type of information by Head Start at the national level.

4.3 Lessons Learned

The findings from this feasibility study provide knowledge relative to several different issues. These are presented below as lessons learned from the study.

Actual Program Practices Do Not Always Fit with Prescribed Program Procedures. While local and national program procedures provided guidelines for how staff recruited and enrolled families, staff sometimes took it upon themselves to assist certain families in the enrollment process, particularly if they truly believed that enrollment was in the best interests of those families. Sometimes this aid took the form of advocating for the family during the enrollment decision process. In some cases, aid took the form of “bending the rules,” such as documenting that a family who really needed Head Start services qualified under the income guidelines, when in fact they may have been ineligible. These activities, however, were not considered gross abuses of the system. Rather, they were presented in terms of sincere efforts to take advantage of opportunities to provide assistance to needy families who otherwise would have not received any assistance at all. It is likely that Head Start rules and procedures were actually maintained during this advocacy process, but across sites, the staff clearly perceived their actions as being at least slightly askew from the norm.

For Many Families, “Need” Is Not Solely Defined by Economics. Head Start eligibility starts with qualifying according to the income guidelines, meaning the program accepts families with incomes at or below the FPL. While families with incomes in this range certainly have economic needs, need is not solely defined by economics. In Love and Grover’s 1987 report on recruitment and enrollment, one of the major conclusions was that the definition of need cannot be limited to economics, as low-income families often had great need in other areas as well. In this study, the staff who participated in the focus groups often hammered home the same point. They reported that they tried to focus on the neediest families by bringing in those facing the greatest challenges. Instead of simple financial hardship, these families also battled other hardships, such as substance abuse, mental illness, domestic violence, or limited education. This is also why staff argued so strongly and eloquently for Head Start to offer services to over-income families: even in better economic conditions, these families were still ‘needy’ and required better links to Head Start and other community services.

The other income-related lesson is that family income is not a fixed value relative to poverty level. The FPL was established as a set of values that vary based on family size, and does not adjust to consider the relative cost of living in different areas of the country. However, families were not only impacted by how well their income could support them in a particular location, they also dealt with variations in local resources. For example, while the cost of living in a rural area may be lower than in an urban area located in the same state, a family in the rural area may face additional transportation limitations and have fewer community support services to assist them. Potential variations in support for local families highlight the need for up-to-date and complete community assessments, so that local Head Start programs can truly be responsive to the needs of their community.

Wide Variations Exist Across Programs in the Management and Use of Waiting Lists. There was a general inconsistency across the local Head Start programs regarding the use and application of waiting lists for enrollment. Although most programs had what they considered a formal waiting list, this list was often formal in name only. While the lists were updated as new families applied and other families already on the list were called to replace dropouts, Head Start staff only verified information for families on the list when actual classroom openings occurred. At that point, it was typically determined that many of the families on the list had already found alternative sources of care or the family was simply no longer able to be contacted. Even though the PIR confirms that classroom turnover is expected each year, there are no guarantees for families on the waiting lists as to if or when they may be offered a slot in the program. Most families needing child care and preschool services for their children will not wait for spaces to open for them. If they are not enrolled in Head Start, they seek and use alternatives where they are available.

National Databases Have Restricted Usefulness for Providing Information on Families Who Are Eligible for Head Start. It was hoped that the use of extant national datasets that were able to identify families with young children as having been enrolled in Head Start would provide useful insights into possible differences between enrolled and non-enrolled Head Start-eligible families. Unfortunately, for the most part, this was not the case. Inherent differences in the construction of the datasets resulted in significant concerns about attempts to make judgements across the datasets. These differences were in areas as basic as the sampling frame and the targeted respondents, and produced variables that were expected to be comparable, but in fact were not because of differences in their definition of terms, including child care, Head Start use, and particularly in the area of family income.

In addition, some of the information about these families that would be most useful to Head Start was not available in these datasets. Such information included further details of family risk (similar to the risks identified by local programs in prioritizing enrollment) and family needs (such as child care) that may impact the ultimate decision to enroll in Head Start. While these datasets represent impressive, national efforts, they were insufficient for the specific purposes of gaining further insights into this study’s target population.

It Is Feasible to Identify and Engage Unserved Families. One of the final efforts of this project was to complete a primary data collection with nine parents of children who were eligible for Head Start, but not enrolled. This task had two components: Identifying the eligible, non-enrolled families, and completing a pilot interview with the primary caregivers in those families.

The study found that three strategies are potentially successful for identifying eligible families. The primary method was the use of listed household samples for each of the two selected locations. As opposed to a list of random phone numbers, the listed samples were targeted to include low-income families and families likely to have children under the age of 5. These samples were applied successfully in the recruitment of families into the study. The use of this general population sample was important in reaching those families not connected to the service community. In addition, parents who were contacted about participation were asked to provide referrals to other families they knew who might be eligible for the study. Most parents responded positively, providing one or two names each. Finally, while in the field to conduct the interviews, the research staff also contacted several of the local service agencies that had participated in the FACES community agency interviews. Without providing individual names, most agencies offered useful suggestions for locating Head Start-eligible families in their service areas, and many offered to actively help contact the families to invite their participation, if needed.

As for the families who were successfully contacted by telephone, all who were eligible to participate indicated they were willing to complete the interview. The interview that was tested during this data collection was similar to that used in the FACES study, but also included a number of sensitive questions about family risks, such as substance use by family members and domestic violence in the child’s home. All inquiries were fully answered without question by the respondents. The tested methods yielded a series of successful contacts and no refusals to participation in the study or to any sensitive questions, suggesting that a larger data collection targeting eligible, non-enrolled families is feasible.

4.4 Head Start’s Future Information Needs

Internal Steps Towards Improved Head Start Information. In terms of information that is available from Head Start, the primary source is the Program Information Report (PIR). Unfortunately, for the purposes of this study, the utility of the PIR was limited because it only provided information at the program level, and not at the center or family level. The Head Start Family Information System (HSFIS) is an important step in generating center and family level information, but it appears that the implementation of this system may not be program-wide, particularly with competing systems, such as Child Plus, already in place. In the future, the ability to capture local data on families who enroll and never show, or who start the program and drop out, may be helpful in identifying needed services for these families. The use of more formal data collection opportunities, such as exit interviews with parents of children who leave the program, could also have great potential for program planning and adaptation to their local communities’ needs.

Feasibility of a Study to Develop an Understanding of Head Start-Eligible, Non-Enrolled Families. The primary data collection efforts for this project offered evidence that families who were Head Start-eligible but unserved were able to be located through the use of targeted telephone lists as well as through referrals from either local agencies or similar families (enrolled or non-enrolled), and that parents identified through these means were willing to participate in interviews or focus groups. The demonstration of feasibility for these two activities suggests that a larger, focused data collection effort would provide nationally generalizable information that was unavailable from existing sources.

Such information is important at this time to the Head Start Bureau. It is anticipated that the Head Start program will continue to expand the numbers of children served. Information sources examined in the present project suggest a significant number of eligible preschool children reside in the communities served by Head Start. Unfortunately, as noted earlier, existing data sources provide little consistent information on a number of important issues, including but not limited to the following:

Certain characteristics of unserved children and their families (family income relative to Head Start eligibility, employment, risk factors encountered by families);
Information regarding preschool programs and other services for low-income families in Head Start communities;
Parental knowledge about and attitudes towards Head Start and other preschool education and child care programs;
Preferred characteristics of preschool programs;
Previous and current need for and use of child care services; and
Parental knowledge and use of other services for low-income families.

Targeting a data collection to families with children under the age of 5 would provide similar information for Early Head Start regarding the availability and use of alternative infant-toddler programs and child care for families eligible for the Early Head Start program. Such information, if available, would provide national and local Head Start administrators and planners with critical information and guidance regarding how these families are currently being served and how they might be served more efficiently. This information would be most useful if gathered from families that were representative of both those with children in Head Start and those with children who were eligible, but not enrolled. Data obtained from such a sample would provide both the necessary information regarding the unserved population and comparisons with current Head Start families.

A national study to obtain the necessary information might take several forms. One would be similar to recent projects undertaken by the Head Start Bureau, including the Descriptive Study of Head Start Health Services, Head Start FACES, and FACES 2000. The central characteristics of this type of study would include:

A nationally representative group of Head Start programs based on information from the most recent Program Information Report dataset. Use of a representative sample of programs and listed household samples would ensure that the unserved families who were located and interviewed would be representative of eligible families residing within service areas of a diverse sample of Head Start programs.
A sample of eligible but unserved families, best located through use of a listed household sample targeted to low-income families within the selected Head Start program service areas. A comparison of current Head Start families from the same locations could be identified as part of the listed sample, but this task would be accomplished more efficiently through collaboration with the local Head Start program. Such a collaboration could also make the identification and participation of program dropouts a consideration. Alternative methods of obtaining samples (referrals from local service agencies) are possible, but would likely restrict the representativeness of the sample of eligible non-enrolled families by not reaching families who are eligible and not participating within the social service system.
A set of in-person interviews would be conducted with eligible but unserved families as well as with Head Start families in the same location. Interviews with local Head Start staff and local community agency personnel could also be conducted during site visits.

A study following the broad design outlined above would provide information necessary for Head Start (and possibly Early Head Start) to adapt, or “improve the fit” between the program and unserved families. Of course, any such modifications should be carefully planned and implemented to avoid reducing the “fit” between the program and currently served families. Access to broad-based information about the characteristics and needs of the additional target families and children is possible, and may enhance the chances that Head Start will meet its long term goals.

This research effort provided much useful information about the Head Start procedures in place for recruitment, enrollment, and retention of families. However, regarding the final answer to the question raised at the start of this chapter, the casual observers of Head Start, the Federal program staff, and the members of the Advisory Committee on Head Start Research and Evaluation will need to encourage further investigation to learn more about eligible families who remain unserved by Head Start.

Table of Contents | Previous | Next